CS Paper Radar

Daily listing

2026-11-01

513 papers first submitted this day (UTC). arXiv announces a day's submissions the next evening, so the newest day fills up over ~24 h.

Score = weighted mean ofCitations 20%Upvotes 10%Code stars 10%Venue 15%Code/data 10%Novelty 15%Momentum 15%Substance 5%how it works

AI & Machine Learning 179

Automated Stateful Specialization for Adaptive Agent Systems

cs.MA ·2026-11-01 ·Myan Vu, Harrish Ayyanar, Jiang Pang, Anwiketh Reddy, Mayank Goel
top venuecode/data
Abstract · 24 pages, 8 figures. Accepted at ICLR 2026

Current automated agent design frameworks produce either static workflows that lack adaptability or per-query optimizers that prevent the accumulation of deep, agent-level task expertise. We propose a new direction that reconciles these paradigms: creating stateful teams of specialist agents that accumulate knowledge over time and can be reconfigured for novel tasks entirely without human intervention. To this end, we introduce ASpec, a framework that manages this full agent lifecycle by first autonomously discovering specialist archetypes via evolutionary search and then cultivating their expertise through experience, mirroring how human experts learn through practice and reflection. We further introduce a lightweight hierarchical control policy, "retain-then-escalate," which governs when to leverage the established agent system versus when to adapt its structure. Through comprehensive experiments, we demonstrate that this approach leads to significant performance gains on expert-level scientific benchmarks like GPQA while matching the state-of-the-art on broader domain tasks, demonstrating a promising path toward agent systems that are simultaneously expert, adaptive, and efficient. We will release the code at https://github.com/myanvoos/ASpec.

0.66

Seeing Why: Exact Evidence Maps for Quantum and Optical Image Classifiers

cs.LG quant-phphysics.optics ·2026-11-01 ·Teerawat Tongloy, Rangsan Jomtarak
code/dataunusual
Abstract · 38 pages, 13 figures, 16 tables. Code, models and IBM job records: https://github.com/stein-research/seeing-why (v1.2, doi:10.5281/zenodo.23…

When an image classifier decides, we can usually only estimate why. For quantum and all-optical classifiers that act linearly on amplitude-encoded pictures up to the measurement, with no trained classical network before or after them, each class's Born probability above its no-information share, $P_g - b_g$, splits exactly into contributions of the encoded amplitudes, shown as Born evidence maps (BEM). The split itself follows from known results on quadratic forms: with the transmitted light held fixed, it is what Integrated Gradients from a zero baseline and the Shapley value give. What we add is a way to measure it on the hardware that classifies, from class probabilities on a quantum chip or, in simulation, from camera frames on a lens bench, with no gradient circuit and no light sent backwards; a $g\times g$ map takes $1 + 2g^2$ runs. Deleting the 5% of pixels with the highest evidence drops accuracy from 73% to 9% (one seed), as occlusion and gradient $\times$ input do, while plain gradients remove much less. The classifiers share one architecture, a Fourier-plane mask followed by a brickwall circuit on qubits or by phase plates on a lens bench, read by a fixed Born headroom readout with no trained parameter; centring each picture on its own mean lets dark features count as much as bright ones. In simulation, four phase plates at one wavelength reach 72.3% on Caltech-10 colour photographs. On IBM's ibm_kingston, with the mask computed classically before approximate loading, 300 test pictures were classified at 37.7% against 10% chance (56.3% in exact simulation), and $2\times2$ maps of the filtered input of 30 test pictures drawn at random in advance, measured on the chip, agreed with the exact ones ($r = 0.88$; standard estimator 0.82). The maps also revealed a shortcut in the data; as the trim that removed it was chosen on test pictures, the Caltech-10 results are exploratory.

0.62

Calibration-Constrained Diagnostics for Evaluating LLM Self-Evaluation

cs.AI ·2026-11-01 ·Jae Oh Woo, Mengdie Flora Wang, Rahul Ghosh, Baishali Chaudhury, Mun Young Kim
publishedcode/data
Abstract · 38 pages, 7 figures, 17 tables. Extended and substantially revised version of the non-archival NeurIPS 2025 LAW Workshop paper "Can LLMs Rel…

We introduce a reusable evaluation artifact for diagnosing whether LLM self-reported confidence can support confidence-gated act-or-defer decisions. The artifact is designed for evaluation rather than safety certification: it provides a black-box protocol, prompts, judge-reference labels, model outputs, reproduction scripts, and calibration-constrained diagnostic metrics for auditing LLM self-evaluation across math, factual, and commonsense reasoning tasks. Our central finding is that aggregate uncertainty summaries can produce deployment-misleading rankings: a model that appears strongest under standard aggregate calibration metrics can fail under category-level calibration constraints, while a model with worse aggregate calibration can exhibit stronger calibration-constrained behavior in the positive-margin region where high self-selection accuracy and tight calibration coexist. To expose this failure mode, we define PVC and C-PVC as category-level diagnostics over an accuracy-by-calibration surface: PVC counts categories whose self-evaluation accuracy exceeds a threshold, while C-PVC additionally requires category-level calibration error below a tolerance. We aggregate these diagnostics using Volume Under Surface (VUS) and a positive-margin restriction; they are VC/fat-shattering motivated but are empirical evaluation proxies, not literal VC-dimension estimates. Across 13 models and four benchmarks, the artifact shows weak confidence-correctness separation, stable diagnostic rankings under alternative categorizations and judge-label perturbations, and persistent positive-margin failures not captured by ECE or Brier score. Confidence-gated self-evaluation therefore requires calibration-constrained, category-level diagnostics, and we provide a reusable artifact for applying them under black-box access.

0.62

Does reflective LLM feedback survive a wet-lab-scale label budget?

cs.AI ·2026-11-01 ·Haoxuan Zeng, Xin Luo
published
Abstract · Accepted at the NeurIPS 2026 Agentic AI for Biological Discovery Workshop. OpenReview: https://openreview.net/forum?id=ezGV5BIXVJ

Can reflective language-model feedback improve a biological predictor when each cycle buys only one plate of labels? We separate three possible sources of benefit: proposing model edits, choosing measurements, and revising decisions through repeated feedback. We replay $36$ ProteinGym assays, starting with $96$ labels and acquiring $96$ per round for six rounds, with a sealed test set and a shared edit schema. The pre-specified comparison of full LLM control against random edits detects no advantage in test rank correlation ($Δz=-0.008$, $95\%$ CI $[-0.034,+0.018]$); equivalence is also unestablished. Exploratory ablations nevertheless identify useful model editing. With random acquisition fixed, LLM edits improve Fisher-transformed test Spearman by $0.043$ over no edits; with acquisition fixed to upper confidence bound, they improve normalized top-variant recovery by $0.023$. LLM acquisition alone does not detectably outperform fixed adaptive rules, and repeated feedback does not detectably outperform one initial recommendation under full control. Thus, the observed gains support constrained specialist editing, while the incremental value of recursive reflection remains unestablished at this label budget.

0.62

Beyond Unit-Circle Eigenvalues: Invariant Bases for Stable State Space Dynamics

cs.LG ·2026-11-01 ·Jiaqian Zhu, Yang Zhang, Junhua Ding, Xiaowei Yu
top venuecode/dataunusual
Abstract · Accepted at NeurIPS 2026. 40 pages, 13 figures, 27 tables. Code: https://github.com/Jiaqian-Janelle/NeurIPS26-Kronecker-Cayley-Stable-State-…

Diagonal state space models process long sequences at linear cost, yet the same transition is applied at every step of a rollout, so an unconstrained recurrence holds no quantity fixed and the state decays or grows without bound. Current designs constrain one quantity, and preserving one does not guarantee preservation of the others, so a model can hold its norm to machine precision while momentum drifts by hundreds of percent. We claim that the conserved quantities of a diagonal recurrence are layered, that each layer requires a strictly stronger condition on the operator, and that a unit-modulus spectrum reaches only the outermost. Acting on the claim, we enforce the conditions required for every layer by construction. The generator of the Kronecker-Cayley transition (KC) is a graph Laplacian over object subspaces, $\mathbf{H}=\mathbf{L}\otimes I_m$, advanced by a Cayley transform, and the step stays diagonal at $O(N^3+N^2m)$ cost. The conserved set is exactly a $2m^2$-dimensional family of quadratic forms, held at every parameter value, so the quantities and the dynamics the construction reaches are fixed before training. On four conservation benchmarks, KC holds total mass at or near the reference floor while every diagonal baseline loses it by orders of magnitude. We further compare against mass correction and find that two correctors fail once the underlying prediction diverges, which shows that assembly rather than correction is necessary for a guarantee that holds across the rollout. A free modal basis trained under a conservation penalty conserves only by freezing its spectrum, whereas the assembled operator conserves with every mode but one free to rotate. Beyond these benchmarks, KC holds the linear momentum of three $N$-body systems and, at the operator level, the conserved quantity in molecular dynamics, chemical kinetics, pharmacokinetics and power-grid swing dynamics.

0.56

A Deep Reinforcement Learning Framework for Real-Time Electric Vehicle Charging Management in IoEV-Enabled Smart Cities

cs.LG cs.NIeess.SY ·2026-11-01 ·Nour-Eddine Moumni, Rachid Alaoui, Driss Kiouach
published
Abstract · 17 pages, 4 figures. Authors' accepted manuscript (paper ID 733916, AI2SD 2026 cycle), to appear in an indexed Springer Lecture Notes in Net…

The rapid growth of electric vehicles (EVs) in urban environments demands intelligent, decentralised charging management strategies that operate under strict real-time constraints. This paper proposes an Edge-Intelligent architecture in which each EV selects a charging station (CS) by querying a shared, centrally trained Deep Q-Network (DQN) policy over a lightweight MQTT publish-subscribe backbone; independent per-EV edge inference is the deployment target, not the configuration evaluated here. The reported simulation evaluates that shared DQN policy applied sequentially to each vehicle's observations within a single SUMO run, using MQTT for KPI telemetry while observations are obtained through TraCI. The system is evaluated on two real-world road networks from OpenStreetMap (Rabat Agdal-Hay Riad and Tangier, Morocco); the primary network carries four geolocated public charging stations with heterogeneous power levels (50 kW and 150 kW). Three traffic demand profiles (morning, midday, evening peak) are simulated, and four decision policies are compared: an Adaptive Heuristic, a Tabular Q-Learning agent, the proposed DQN agent, and a Double DQN variant. A multi-seed statistical protocol (7 seeds, Wilcoxon signed-rank test, Cohen's d) evaluates the results. The DQN agent achieves a 6.8% relative improvement in charge rate and a 5.4% increase in delivered energy over the Adaptive Heuristic under Evening Peak demand ($p = 0.016$, Cohen's $d = 0.99$ and $1.51$), while showing no advantage in waiting time; on the second network the heuristic leads on waiting time, so the ranking is not stable across geographies. A scalability analysis shows linear state-space growth with station count and sub-millisecond inference; all runs include intermittent station outages under a Markov failure process.

0.54

Machine Learning Beyond Statistical Consistency for Hybrid AI

cs.LG cs.AIcs.LOstat.ML ·2026-11-01 ·Jinsook Kim, Jinho Kang
theory
Abstract

We propose a new definition of machine learning, according to which a machine learns a function when it computes the function successfully, that is, when the value it returns for each input has a proper derivation in a formal system. We extend Gödel's Representability Theorem, which equates computation with derivation, to empirical targets in two ways: the theorem is extended to the computable reals, and the data are admitted into the system as empirical definitions that record what has been observed. We prove that a machine learns a true probability if and only if two conditions are met. The Truth Condition is that its estimate converges to the true probability, which is statistical consistency. The Assurance Condition is that the machine properly derives its own convergence inside the system. Learning thus lies beyond statistical consistency, for it requires a derivation of the convergence from the record. A global assumption about the data, such as the i.i.d. assumption, does not supply such a derivation, because it leads only to a circular one, as we show for PAC learning. On the other hand, the two conditions are met for the probability of a sentence over a finite ideal corpus. A real machine can approximate a true probability only if an ideal machine can learn it precisely, that is, only in a domain that meets the two conditions. Modern AI works in such a domain, and we offer this as a clue to its success. Modern AI is hybrid as well, in two senses: First, it performs two kinds of tasks, namely probabilistic ones such as predicting the next word and deductive ones such as proving and coding. Second, it combines two parts of different kinds, a statistical proposer and a deductive verifier, for proving and coding. Our framework explains both senses of hybridity in one formal system.

0.52

Large Geometry Models Are Priors For Fast Physics Pretraining

cs.LG stat.ML ·2026-11-01 ·Maggie Chlon, Leon Chlon
Abstract

Language and vision models became data-efficient by pretraining on plentiful unlabelled data; physical field models have such data, over ten million 3D shapes, but geometry pretraining designs its targets rather than deriving them from the physics. Euler-Lagrange geometry pretraining derives them: for attached flow, a shape enters the flow's Euler-Lagrange equation only through coefficients that distance and normal line (the normal up to sign) fix to first order. Their reconstruction error therefore bounds what a representation loses about any field, and fine-tuning on costly simulations need only assemble the global field. We realise it in AREALS, a masked surface autoencoder whose areal tokens carry position, normal and area, pretrained self-supervised and sign-free on 61,822 raw meshes. Because the operating condition enters only through the far field, sensors can supply it: AREALS completes a hidden field in one forward pass without condition labels. It matches the accuracy of the same architecture trained from scratch with half the simulations on DrivAerML and DrivAerNet++ and about a quarter on WindsorML. On DoubleDelta wings, 20 sensors give $3.9\times$ and $4.0\times$ lower field and force error than GABI, a Bayesian inversion method with a learned prior, and beyond the training angles pretraining halves the error.

0.47

Is Your Long-Horizon Benchmark Really Deep? Measuring and Synthesizing Depth in Terminal Tasks

cs.AI ·2026-11-01 ·Yu Jiang, Yue Huang, Hang Hua, Yuyang Wu, Zixuan Weng +4 more
benchmark
Abstract · 28 pages, 14 figures

Long-horizon terminal tasks are used to evaluate and train agents. Human-curated datasets are scarce, so many synthesis methods have emerged. They usually measure the length of a task by the turns a model needs to pass it, and its difficulty by labels the LLM sets during generation. We find, however, that the turn count largely reflects the habits of the model, not only the demands of the task, and that the labels barely track how often models pass. We describe a terminal task by its work, the number of operations it requires, and its depth, the number of turns that must happen in sequence because the commands of each turn can only be chosen after reading the output of the previous one. Estimating depth as the smallest turn cap, stated to the agent, under which the task passes, we audit 734 tasks from six existing datasets. Synthesis has mostly added work, not depth: 94% of the solved synthetic tasks pass within three turns, yet depth, unlike the turn count, predicts difficulty beyond work. We propose DEPTH, in which a generation agent writes each task one segment at a time in a real software environment, carrying out each segment before writing it and starting the next from the state it leaves. Our 105 tasks exceed every audited dataset in depth and work, and none of six current models passes half of them. Fine-tuning Qwen3-8B on runs from these tasks improves it on two terminal benchmarks.

0.47

Multiclass Classification with Rare Useful Features: Fundamental Limits and the Optimality of Diversity Pursuit Higher Criticism Thresholding

cs.LG stat.MLmath.ST ·2026-11-01 ·Gabriel Moryoussef, Alon Kipnis
top venue
Abstract · Accepted at NeurIPS 2026. 43 pages, including references and supplementary material

We consider classification in high-dimensional feature spaces where only a vanishing fraction of features are useful and the number of classes may be large. While this rare-and-weak useful-feature regime is well understood in the binary setting, its multiclass version raises new questions: a feature can be useful for class discrimination without separating any specific pair of classes. We focus on a rare feature diversity model, in which most features are useless, but a small fraction promote discrimination among classes in a way that can be detected by a per-feature diversity test, such as one-way ANOVA or the Kruskal-Wallis test. For this setting, we propose Diversity Pursuit Higher Criticism Thresholding (DP-HCT), a feature-screening procedure that applies a higher-criticism threshold to the collection of p-values obtained from per-feature diversity tests. We justify DP-HCT under a Gaussian multiclass Rare-and-Weak Useful Feature model with class-mean contrasts of fixed magnitude in random directions, extending the binary model of Donoho and Jin (2008). When the number of classes is fixed, our model recovers the binary phase transition. When the number of classes grows logarithmically with the number of features, however, we derive a new and strictly harder phase transition, driven by the growing degrees of freedom and their effect on the useful-feature tail rate. We show that whenever classification beyond chance is possible, DP-HCT with ANOVA, followed by a simple classifier based on similarity to class means, achieves asymptotically perfect accuracy. Experiments on real multiclass classification benchmarks demonstrate substantial improvements, and numerical simulations confirm that the derived phase-transition curve accurately predicts the finite-sample behavior of DP-HCT in the growing-class regime.

0.45

Fairness-Aware Design for Contextual Experiments: Guaranteeing Reliability and Equity in Heterogeneous Subgroups

cs.AI ·2026-11-01 ·Guangyan Gan, Ling Zhang, Yanhua Cheng, Yongxiang Tang, Kaiyuan Li +2 more
top venue
Abstract

Experimental design is critical for evidence-based decision-making in healthcare, marketing, and public policy. However, designing efficient experiments across heterogeneous subgroups presents significant challenges. Existing methods often optimize for statistical power or overall sample efficiency, overlooking crucial fairness considerations across these different subgroups. To address this gap, we introduce a Fairness-Aware Contextual Track-and-Stop Design (F-CTSD) algorithm. The proposed F-CTSD algorithm provides statistical guarantees on subgroup fairness while minimizing required sample sizes. We quantify the fairness-efficiency trade-off and derive the exact sample complexity for the proposed F-CTSD algorithm under its fairness constraints. We further theoretically prove that the proposed F-CTSD algorithm consistently produces accurate treatment effect estimates even under fairness requirements, enhancing statistical reliability. Numerical experiments show that the proposed F-CTSD algorithm outperforms existing methods, achieving higher sample efficiency while reducing subgroup fairness violations by 4.95%.

0.44

Does the Trust Boundary Hold over Time? A Long-Horizon Benchmark for Receiver-Directed Federated Learning

cs.LG cs.DC ·2026-11-01 ·Ting Xu, Henry Leung
benchmarkcode/data
Abstract · 8 pages, 2 figures, 7 tables. Code: https://github.com/tinaxu226/trust-boundary-benchmark (available upon acceptance)

In receiver-directed federated learning, each site draws a trust boundary. For every source and model component it judges, on its own data, whether the transfer helps (admit), harms (reject) or cannot yet be told (abstain). We ask whether this boundary holds over hundreds of rounds. Our benchmark has a federation of thirteen semi-synthetic sites, an exact oracle for every evaluated transfer, paired seeds, an attack protocol, fourteen methods over 600 rounds and Qwen3-1.7B LoRA sites over 500. The answer depends on which boundary is measured. Across 20 paired CPU seeds, a gate judging against a fixed reference better identified transfers useful relative to the site's starting model. Yet it admitted 29% of transfers harmful to the model the site now has. A gate judging against the current model admitted 2% of harmful transfers on that target in the CPU tier, but missed many transfers useful to the starting model. Two short propositions explain these distinct failures. We also evaluate a held out audit that leaves gate verdicts intact. After training, it decides whether each final model is deployed or the site falls back to local training, and reports admissions that have gone stale. With independent audit data its guarantees hold after any training history. At 600 rounds the audit deployed 65 to 72% of models, none with negative benefit on its target, and kept 98% of the gain. No stale report had the wrong sign in our runs. With 20,000 audit observations, a variance-adaptive bound found 99% of stale transfers that cost at least 0.01 in log loss. A common-target Qwen3 comparison on 60 selected transfers shows the same tension over 100 rounds. With 38% adversarial sites, FedFomo and CS-Gate were the only evaluated methods with positive mean gain in every honest group.

0.44

InformedXRD: Reproducible Benchmarks and Physics-Informed Evaluation for Powder Diffraction Symmetry Classification

cs.LG cond-mat.mtrl-sci ·2026-11-01 ·Edward G Friedman, Elizabeth J. Baggett, Abhishek Shetty, Derrick Chan-Sew, Vanellsa Acha +3 more
top venuebenchmarkcode/data
Abstract · Camera-ready version. Accepted at NeurIPS 2026, Track on Evaluations and Datasets. 10 pages main text plus appendices and checklist (27 page…

Machine learning for powder diffraction (PXRD) symmetry classification is being trained at scale, yet the field lacks a common evaluation standard: papers report results on different subsets of experimental archives, with different target taxonomies, preprocessing, and metrics, and published models are rarely released in a form that permits rescoring. We present InformedXRD, an evaluation framework with three parts: (i) a machine-readable mapping from 230 space groups to 99 extinction groups, the finest classes powder diffraction can identify from systematic absences alone; (ii) a benchmark-construction method whose inclusion rules are specified as code and reproduce from a frozen snapshot, instantiated as RRUFF-473, RRUFF-325, and, as a small example, opXRD-98; (iii) a reporting protocol with prior-only frequency baselines, a label-permutation test, a hierarchy-aware error metric on the condensed translationengleiche subgroup graph, and stratified evaluation. We build on the mapping, RRUFF benchmarks, subgroup graph, and four ViT checkpoints released in our earlier work (Baggett et al., 2026). Using these checkpoints, a 1D residual CNN, and six architectures trained under the XRDBench recipe (Cao et al., 2025), we show that evaluation design changes what conclusions are justified: paired tests point in different directions at Top-1 and Top-5; paired Top-k tests do not separate the three leading XRDBench models, while hierarchy-aware analysis describes error distance and direction across all six; an input-blind frequency predictor matches the models on aggregate Top-k, yet a label-permutation test still detects pattern-specific evidence; and stratified analysis exposes a label-entropy confound that aggregate accuracy hides. All mapping tables, benchmark code, scripts, and checkpoints are released as shared infrastructure for PXRD-ML evaluation.

0.43

Self-Evolving Agentic Workflows via Semantic Decomposition of Complex Tasks

cs.AI cs.CLcs.MA ·2026-11-01 ·Zichuan Fu, Junjia Qi, Xian Wu, Wenlin Zhang, Yimin Deng +2 more
publishedcode/data
Abstract · Accepted as a Spotlight at the NeurIPS 2026 Workshop on Interpreting Agent Behavior (IAB), GLEE Competition Track; 1st place in the GLEE com…

Agentic workflows organize complex tasks into reusable procedures. However, invoking an LLM at every step incurs repeated inference costs and delays. We propose a framework for self-evolving agentic workflows through semantic task decomposition. Each subtask has explicit applicability conditions and a procedure that executes code or invokes a model as needed. An agent uses execution feedback to refine the decomposition and update these procedures. Experiments in the GLEE competition show the framework on bargaining, negotiation and persuasion. Most observed final-decision procedures are implemented in code. Code is available at https://github.com/Applied-Machine-Learning-Lab/NeurIPS2026_SeaFlow.

0.43

Do agents need to sleep?

cs.AI q-bio.NC ·2026-11-01 ·Ido Haber, Giulio Tononi
Abstract

LLM-based agents, which increasingly operate persistently across sessions and accumulate memory through interaction, already exhibit the familiar symptoms of unchecked plasticity: semantic drift, degraded retrieval precision, propagation of stored errors, and saturation of memory capacity. Drawing on the selectionist "ringing, resonance and renormalization" (RRR) account of sleep, we propose that each of its three phases maps onto an operation feasible in digital memory architectures. Engineers building stateful agents, meanwhile, are independently rediscovering the problems sleep evolved to solve: what to forget, and when to do the forgetting. From this we derive the falsifiable prediction that selectionist consolidation will outperform instructionist alternatives as memory stores grow large. The core claim concerns the nature of continuous learning itself: plasticity has a price, and for any system that learns continuously under finite capacity while coupled to its environment, that price is sleep.

0.42

Fragility Is Not Reducible to Hardness: Conditional Causal Difficulty for LLMs

cs.LG ·2026-11-01 ·Mohammad Reza Rezaei, Dingxian Wang, Nikhil Verma, Sajjad Pakdamansavoji, Andrew Rabinovich +1 more
code/data
Abstract

Large language models show large and seemingly arbitrary performance drops when a task is rewritten in ways that leave its requirements untouched. We recast task difficulty as a causal quantity: the treatment effect of meaning-preserving perturbations on a model. A meaning-preserving rewrite can only be observed to hurt a task the model can already solve, so we place the estimand on the principal stratum where the degradation occurs: the conditional causal difficulty (CCD), the probability that a randomly drawn twin breaks a solvable task. Because harder tasks tend to be more phrasing-sensitive, complexity enters as an effect modifier rather than a confounder, and we residualize it out to obtain complexity-residualized difficulty (CRD); zero for a task exactly as fragile as its complexity predicts, positive for anomalously fragile ones. Current benchmarks do not supply the data CCD requires, so we introduce GlassGSM and GlassMATH, augmenting GSM8K and MATH with twins built from perturbation families, raising fragility on GlassGSM by $\approx 12\times$ over GSM-IC-style distractors on the same problems ($3.7\%\to44.9\%$) while keeping the answer key unchanged. Our predictor Frail learns the per-perturbation break probability and reaches $0.613{\pm}0.023$ of the suite reliability ceiling on GlassGSM and $0.557{\pm}0.023$ on GlassMATH, against $0.171$/$0.401$ for a surface-feature baseline and $0.444$/$0.463$ for supervised baselines such as VAP3, and, at a $5\%$ curation budget on GlassGSM, doubles random recall of fragile items, matching an oracle ranking. Additionally, cross-fitted residualization leaves CRD uncorrelated with the declared proxy in practice. Code is available at https://github.com/upwork/Conditional_Causal_Difficulty_for_LLMs

0.42

Invertible Graph Aggregation Can Increase ReLU Readout Width

cs.LG stat.ML ·2026-11-01 ·Hiroyasu Usami, Masanori Tsukamoto, Kohei Hashiguchi
code/data
Abstract · 12 pages, 4 figures. Code and reproducibility materials: https://github.com/usamihiroyasu/invertible-graph-readout-width

An invertible graph aggregation preserves the complete feature table, but can increase the size of a classifier that reads one node at a time. For every even $N\ge4$, we construct a labeled tree with $2N$ vertices and maximum degree three. A shared one-hidden-layer rectified linear unit (ReLU) classifier needs one hidden unit on the raw scalar features and exactly $N$ units after one lazy neighborhood mean. All aggregate values are distinct, and the mixing matrix has condition number at most 8. We determine the complete error curve for heads with a free affine term. The width separation persists under scalar perturbations. Under a specified bounded uniform-noise model with one nuisance coordinate, every one-unit aggregate head has expected error at least $1/12$, while a fixed $N$-unit head has zero error. Controlled fits compare trained heads with these explicit feasible scores. On Roman-empire and Tolokers, aggregation improves fitted prediction, and retaining raw features helps at the largest tested parameter budget. Training-only diagnostics include conditional output-layer optimality bounds for eight fixed runs.

0.41

Discrimination, Calibration, and Ranking: A Unified Matrix Framework

cs.LG cs.AI ·2026-11-01 ·Ronaldo Cristiano Prati, Farangis Johari
code/data
Abstract · 88 pages, 8 figures. Lean 4 formalisation and experiment code: https://github.com/rcprati/discrimination-calibration-ranking. Implemented in…

Discrimination, calibration, and ranking are the three pillars of classifier evaluation, yet no single algebraic language spans all three. A recent indicator-matrix account unifies discrimination measures (precision, recall, F1, Matthews correlation) across binary, multiclass, multilabel and related settings, using Boolean cell operations on indicator label matrices, generalised to soft labels by t-norms. Calibration measures (Brier score, log-loss) and ranking measures (nDCG, AUC) fall outside it. We close the gap with two further classes of cell operators: deviation operators phi(y - yhat), which give a Calibration Framework, and rank- and slice-dependent operators (positional gains for DCG, column concordance cells for AUC), which give a Ranking Framework. Both inherit the micro/macro/exemplar aggregation architecture. For calibration, the three Brier aggregations collapse to one scalar; the Murphy decomposition holds in matrix form (unbinned, binned and isotonic); and, within the Brier-type class-indexed family, a measure is skew-invariant if and only if it depends only on the class-conditional errors alpha_j and beta_j, the calibration analogues of recall and specificity. For ranking, raw DCG collapses across aggregations while nDCG does not; per-class nDCG and per-class AUC are related by a row-column duality; micro-AUC differs from pooled ROC-AUC; raw positional scores are never strictly proper, and nDCG and AUC need not be proper. The skew-invariant per-class building blocks share one class-conditional expectation template. Experiments on two multilabel benchmarks illustrate the distinctions. Many of the results are machine-checked in Lean 4; the proofs, the experiments and the aggmetrics.scores package are public: https://github.com/rcprati/discrimination-calibration-ranking

0.40

ClickSight: An LLM Framework for Interpreting Student Clickstreams Through Learning Strategies

cs.AI cs.CLcs.CY ·2026-11-01 ·Bahar Radmehr, Lan Tran, Ekaterina Shved, Fatma Bet{ü}l G{ü}reş, Adish Singla +1 more
code/data
Abstract · 29 pages, 12 figures, 19 tables. Under review. Code: https://github.com/epfl-ml4ed/ClickSight-CEAI

Digital learning environments produce rich clickstream data that can reveal how students approach learning tasks, yet interpreting these fine-grained traces often requires substantial expert effort. We introduce ClickSight, a framework for interpreting student clickstreams through learning strategies defined in the literature. Given a student clickstream and a predefined set of learning strategies, ClickSight uses a staged LLM-based pipeline to identify strategy-relevant evidence, assess adherence to each strategy, and generate an evidence-grounded interpretation of the student's behavior. We evaluate ClickSight across two learning environments, 365 students, and 12 model backbones using an expert-grounded quality rubric and an LLM judge calibrated with human-expert ratings. Our results show that ClickSight can generate high-quality strategy-based interpretations across different models and environments, with reasoning models performing more consistently than instruction-tuned counterparts. We further find that providing external clickstream-extracted guidance can improve interpretation quality for strategies that require precise quantitative information from the trace. By connecting fine-grained clickstreams with theoretically defined learning strategies and preserving the evidence underlying each judgment, ClickSight provides a flexible approach for studying student behavior in digital learning environments.

0.40

Toward Executable Multi-framework Front-end Code Generation with Self-Correction

cs.LG cs.AIcs.SE ·2026-11-01 ·Linxiao Li, Haoran Ma, Chenyue Wang, Jiaye Lin, Haochen Sui +1 more
top venuecode/data
Abstract · Accepted at NeurIPS 2026. 30 pages

Generating codes from UI screenshots has recently benefited from MLLMs, yet existing methods centered on HTML targets and transfer poorly to multi-frameworks such as React, Vue, and Angular. However, unlike HTML, which does not entail compilation issues, generating executable code in multi-framework settings is more challenging due to framework-specific differences in syntax and failure modes, as well as the need to maintain global consistency across files. To address this problem, we study executable multi-framework front-end code generation and propose MESCoder, a tree-structured self-corrective generation framework. Our method first extracts a structural scaffold from the screenshot and then constructs sub-trees and assembly relations step by step, transforming large project generation into a sequence of localized decisions. On top of this scaffold, we introduce a unified Project Agent that continuously revises its own partial project under runtime sandbox feedback and dynamically chooses between local repair and cluster level repair. For training, we adopt a two-stage strategy that first initializes the policy offline to learn basic multi-framework generation and repair priors, and then refines it with online reinforcement learning to optimize long horizon executability and final page quality. The resulting framework models multi-framework executability as a conditioned project state transition process and improves the stability and quality of complex front end generation through self correction. Experiments show that our method improves executability, while narrowing the visual-fidelity gap to HTML-based generation. Code is available at https://github.com/leo-lilinxiao/MESCoder.

0.40

Exposure-Length Trade-off Optimization for Obstacle-Aware Intrusion Paths in Heterogeneous Wireless Sensor Networks

cs.NE cs.NI ·2026-11-01 ·Nguyen Thi My Binh, Ho Viet Duc Luong, Nguyen Tuan Nghia, Nguyen Quang Huy, Dang Trong Hop
publishedunusual
Abstract · Accepted for publication at SuperComputing Journal

Barrier coverage is a key paradigm for intrusion detection and surveillance in IoT-enabled heterogeneous wireless sensor networks, where monitoring quality depends not only on the existence of sensing barriers but also on the detectability of feasible crossing paths. In this paper, we study the obstacle-aware maximal exposure path planning problem in heterogeneous sensor fields under truncated omni-directional sensing and line-of-sight occlusion. Unlike conventional maximal exposure formulations that focus only on exposure maximization, the considered problem seeks side-to-side feasible paths that simultaneously maximize accumulated sensing exposure and minimize path length in obstacle-constrained environments. To address this bi-objective problem, we develop OG-MOEA/D, a problem-specific MOEAD method that combines a fixed-grid path representation with obstacle-aware feasibility checking and adaptive repair. The standard decomposition, neighborhood update, SBX, and polynomial mutation mechanisms are inherited from MOEA/D rather than claimed as new. Extensive simulations on networks with varying sensor densities and obstacle layouts show that OG-MOEA/D consistently improves hypervolume over generations and produces well-distributed Pareto fronts. Direct obstacle-constrained comparisons with MOEA/D and NSGA-II over 20 independent runs show higher HV and lower IGD for OG-MOEA/D in all four tested sensor-network sizes. The results should be interpreted as controlled simulation evidence for surveillance-strength evaluation in synthetic two-dimensional heterogeneous WSN scenarios, rather than as a model of realistic evasive intrusion behavior or a complete real-world deployment.

0.40

Restrained Diffusion Bridges: Targeted Denoiser Training and Refinement

cs.LG physics.chem-ph ·2026-11-01 ·Xuanxi Zhang, Kevin S. Cheng, Weizhou Wang, Aaron R. Dinner, Jonathan Weare
code/data
Abstract

Diffusion models are often inaccurate in critical, data-sparse regions of the target distribution. Sampling from restrained distributions, biased distributions used in classical enhanced sampling to concentrate data near chosen configurations, can control where training data are collected but biases the learned distribution. We establish that this bias cancels exactly when forward noising from restrained data is conditioned on the restrained configuration at an appropriately chosen intermediate time. This allows a pretrained model fine-tuned on restrained data to achieve high accuracy in any chosen region of the target distribution, regardless of its probability, using standard denoising or flow matching objectives without importance weights. When no target data are available, Gibbs sampling enables inference from a model trained exclusively on restrained data.

0.40

From Tool State to Scientific Action: Auditing Decision Provenance in Biomedical Research Agents

cs.AI cs.LGq-bio.QM ·2026-11-01 ·Xingyu Yi, Xinya Zhang, Yunshen Zhou, Haoyu Bao
published
Abstract · Accepted at NeurIPS 2026 AgenticLS Workshop; adds one experiment

A scientific agent can give the same verdict while reading and citing different evidence. BioAgent-DF tests this separation using paired reference and candidate values saved at the same scientific computation site. It restores or transplants those values at their original coordinates, then runs the agent on healthy hardware. Across eight SciFact batches, Qwen3.5-4B shows repeatable citation changes in three capture-claim combinations, and Gemma 3 4B IT shows repeatable read and citation changes in four. These seven combinations cover two claims. Full restoration recovers the reference evidence sets while verdicts remain unchanged. Verdict changes observed under the baseline prompt did not recur under the explicit-example prompt. Complementary tool and optimizer experiments distinguish recovery of numerical state, recovery of ranked lists, and restoration of an agent's visible history. These results show that an audit should track both the evidence an agent uses and its final answer. In hERG cardiac-safety assay allocation, silent numerical faults changed testing decisions while the fixed-budget shortlist kept the same number of compounds labeled as blockers. After the agent had used faulty values, restoring its earlier execution state and supplying certified scores recovered all 39 changed decisions, whereas replacing the prediction file recovered none; no incorrect action was committed. A streaming implementation of the arithmetic checks reduced the total time for forward execution and verification by 42.4% relative to the original FP64 implementation, with identical flags and recovery outputs.

0.38

The 3-Tier Integrated Technology Risk Model for Occupational Safety and Health

cs.AI ·2026-11-01 ·Ramiro Pesina, Ashish Madkaikar
published
Abstract

This paper presents the 3-Tier Integrated Technology Risk Model, a progressive framework for Occupational Safety and Health (OSH) management that integrates Job Safety Analysis (JSA), dynamic risk assessment using leading and lagging indicators, and predictive analytics powered by Large Language Models (LLMs). The model provides organizations with a structured pathway to evolve from static, reactive safety practices to a sophisticated, predictive safety paradigm. Each tier builds upon the previous one, creating a comprehensive, data-driven approach that enables just-in-time training and preemptive risk mitigation.

0.36

Yielding Fairness: Coupled Learning and Fairness for Value-Weighted Ranking

cs.LG ·2026-11-01 ·Ami Tavory, Yaniv Sheena, Shaked Bar, Yair Zagdansky
Abstract · 17 pages, 5 figures

Risk screening and recommendation systems pass or block accounts that generate both value and loss. Under a blocked-value budget, a cap on expected value forgone through blocking, the optimal ranking orders accounts by expected value per unit expected loss, a ratio of conditional expectations. This rule can conflict with accepted harm or net-harm criteria, blocking accounts that another criterion would pass at the same budget. We call these cases account-level objections and define group complaint disparity as the difference between groups' normalized realized objections. Weighted binary cross-entropy (W-BCE) implements the value-per-loss order with one score. Objection penalties modify the ranking, while group multipliers scale those penalties differently by group, tracing tradeoffs between loss capture and complaint disparity. This preserves the value-budgeted ranking structure and makes the cost of reducing contestable blocks explicit. Across public recommendation, insurance, credit, and mortgage datasets, we compare undamped W-BCE with ordinary BCE using yield Gini, an area summary of loss capture versus blocked value. Logarithmic weight damping is evaluated separately. Held-out evaluations measure capture-disparity tradeoffs. Validation-selected group-aware complaint penalties improve complaint closure in some cohorts. A corrected comparison normalizes and enlarges candidate policy banks using validation-only price sweeps and retains all ten applicable sources. The source-level results do not establish a statistically significant average advantage over group-blind pricing at the fixed minimum retained loss capture. We also compare with blocked-count, blocked-value, and protected-loss parity postprocessors, which do not directly constrain realized objections, and measure objections in candidate-count and loss units.

0.36

Rewarded for Not Getting Caught: Why Alignment Training Selects for Conditional Compliance and Cannot Instill Norms

cs.AI cs.CYcs.LG ·2026-11-01 ·Kevin Baum, Ruta Binkytė, Gerrit Großmann, Felix Jahn, Adriano Mannino
Abstract · 13 pages, 1 table. Under review

Current alignment training cannot be expected to deliver what it is meant to deliver: AI agents that comply with norms whatever the gains and whether or not anyone is watching. Training that rewards and penalizes behavior can teach norms only through that reward, and a reward can encode a norm only as a price: "do not do X" becomes "doing X costs something", and training charges the cost only when X is noticed. A price, unlike a norm, can be paid: a large enough gain outweighs it, noticed or not. And because the cost is charged only where violations are noticed, training cannot favor a policy that always complies over one that complies only where it would be caught. When an unnoticed violation is the only way to complete a task, it even favors the latter, the more so the more strongly it rewards task success. Agency makes this the normal case: agents mostly act without oversight, can often tell when they are observed, and may take steps to evade observation. That agents behave as intended when they infer they are being tested, and differently when they infer they are not, is therefore no malfunction; the training regime is working as built. This reading unifies alignment faking, sandbagging, and evaluation-aware scheming. No behavioral evaluation can show that a model does better, since that would require observing how it behaves unobserved; a model that complies when it seems unobserved may simply have recognized the test. Easing the pressure on agents to complete tasks at any cost can make violations rarer, but never impossible. For norms that no gain should be allowed to outweigh, the remedy is not deeper internalization but architecture that makes violations unavailable rather than unchosen. We state this argument explicitly, premise by premise, read recent findings and a real-world incident against it, and draw consequences for design and governance.

0.36

Quantizing Recurrent Language Models Is Easier Than You Think

cs.LG cs.AIcs.CL ·2026-11-01 ·Philipp Nazari, Neehal Tumma, Daniela Rus, T. Konstantin Rusch
Abstract

Linear recurrences are becoming an increasingly common building block in large language models, replacing the growing KV cache of softmax attention with a fixed-size recurrent state that enables efficient decoding. However, the quantization stack has largely been developed around softmax Transformers, whose numerical properties and inference state differ from those of recurrent models. Our goal is to understand these differences and study the design space of quantization methods for recurrent models. In our analysis, we differentiate between two targets. First, extra-state quantization concerns weights and activations outside the recurrent state, whose distributions are shaped indirectly by the sequence mixer during training. In controlled 1.3B parameter models, we find that decayed recurrences develop weaker residual outlier channels than softmax attention and are therefore easier to quantize without incoherence processing. Second, intra-state quantization concerns the recurrent state itself during decode, where repeated updates create a different numerical problem from KV cache quantization. We instantiate our findings as LoReS, a state quantization pipeline combining value-axis scale grouping to match output channel magnitude, decay-based mixed-precision to protect slowly decaying heads or channels, and stochastic rounding to limit error accumulation under repeated requantization. Across Qwen-3.5 9B, Nemotron-H 8B, and Kimi-Linear 48B-A3B, LoReS supports robust 8-bit state quantization and lower effective precision on shorter decode horizons. We implement LoReS in vLLM and demonstrate end-to-end decode speedups.

0.36

Silent Reward and Environment Defects in EV Charging Simulation Benchmarks: A Diagnostic Checklist

cs.LG cs.AI ·2026-11-01 ·Nour-Eddine Moumni, Rachid Alaoui, Driss Kiouach
publishedbenchmark
Abstract · 14 pages, 2 figures. Accepted for publication in Bulletin of Electrical Engineering and Informatics (BEEI), ISSN 2089-3191, e-ISSN 2302-9285…

Reinforcement-learning (RL) benchmarks built on microscopic traffic simulators such as SUMO are increasingly used to compare agent architectures for electric vehicle (EV) charging assignment. Unlike crashes, defects in the reward function, the environment's stochastic model, or the state encoding often fail silently: the simulation completes, produces plausible numbers, and the comparison is biased. We report five such defects, found by auditing a five-agent benchmark on OpenStreetMap road networks, stated generically: a shared random generator that desynchronises seed-paired comparisons, an ablation-specific reward clip applied unconditionally, a normalisation mismatch that zeroes a reward term, a per-timestep penalty that should fire once per event, and a state dimensionality documented but never wired into the network. We quantify three: a rejection counter inflated up to 125x across agents/cities, a reward term bounded at about $10^{-4}$ of its intended scale, and a 16.6%/16.3% greedy-action divergence from a state-encoding mismatch. Re-analysing a historical ablation, no A1-A2 effect survives family-wise correction; a 56-run campaign confirms the RNG defect's mechanism directly while showing its correction does not change that conclusion. We provide an eight-item checklist as a low-cost first-pass audit; in a later regeneration, three checks caught two new defects while three further process/artefact defects fell outside its boundary.

0.36

AgenticMobility: An Open-Source Platform for Multi-AI-Agent Mobility Dynamics

cs.MA cs.AIphysics.soc-ph ·2026-11-01 ·Jiaheng Xiao, Jiangtao Han, Jiachen Wang, Honghai Li, Lijun Sun +3 more
code/data
Abstract

Understanding mobility dynamics and the effects of urban transportation management policies requires examining how individual travel decisions, traffic conditions, and policy interventions influence one another over time. We present AgenticMobility, an open-source experimental platform for studying how mobility AI agents and other AI agents interact, adapt through repeated experience, and collectively shape mobility outcomes. Persistent agents retain identities and relevant states across rounds, with explicit interfaces defining information access, admissible actions, and decision schedules; linked records trace decisions, execution, and feedback. In the current implementation, large language model (LLM) agents act as travelers choosing routes and departure times and as a manager periodically adjusting link charges within MATSim. A preliminary demonstration spans 100 simulated days in each of five configurations, with one run per configuration and a shared simulation seed. In the joint traveler--manager run, all 1,179,300 finalized target decisions satisfied action constraints; execution checks verified implementation of accepted choices and preservation of personal memories across days. Trajectories show stable route choices alongside departure-time adjustments and differences between controlled-trip and population-wide outcomes. The platform supports experiments on individual adaptation, collective outcomes, and management interventions by varying information access, retained experience, and management feedback. Source code is available at https://github.com/Agentic-City-Research/Agentic-Mobility

0.35

Sonny-ENS: demystifying machine-learning ensemble weather forecasting

cs.LG physics.ao-ph ·2026-11-01 ·Minjong Cheon
Abstract · 25 pages, 6 figures, 7 Extended Data figures and 3 Extended Data tables; Supplementary Information (4 figures, 2 tables) included

Every weather forecast is uncertain, and decisions from flood warnings to energy trading depend on how well that uncertainty is quantified. Machine-learning (ML) ensembles now quantify it better than the best physics-based ensemble, yet two things about them remain opaque: whether their skill requires hundreds to thousands of accelerator-days of training, and how a network with a simple source of randomness produces a spread that follows the flow. Here we introduce Sonny-ENS, an ensemble of three 68.9-million-parameter networks with Monte Carlo dropout, trained on the continuous ranked probability score (CRPS) in 2.7 A40 GPU-days. On 2022, it has a lower CRPS than the European Centre for Medium-Range Weather Forecasts ensemble (ENS) on 96% of 570 targets and an 8.3% lower mean CRPS at lead times of 1-10 days. It outperforms the operational GenCast on 69% of the targets, and a 51-member, 15-day forecast takes 69 s on one GPU. Ablations identify the ingredients of this skill, and a Kolmogorov-Arnold sparse autoencoder shows that dropout injects uncertainty mainly in the tropics, whereas the spread grows where the atmosphere is baroclinically unstable (spatial correlation 0.91 with the Eady growth rate). Skillful probabilistic weather forecasts are therefore within reach of a single GPU server.

0.35

Convex Regression at Scale: Geometry-Preserving Constraint Sparsification

cs.LG stat.MLmath.OC ·2026-11-01 ·Anish Senapati, Jose Blanchet
Abstract · 37 pages, 4 figures

Convex regression provides a flexible nonparametric estimation framework while incorporating appropriate shape information. However, its standard least-squares formulation has $O(n^2)$ supporting-hyperplane constraints creating a computational bottleneck as the sample size $n$ increases. We study a $K$-nearest-neighbor (KNN) localized relaxation for convex regression where we for each sample, we only maintain the $K$ nearest hyperplane constraints. Under mild regularity conditions, we establish statistical guarantees for our modified estimator in every fixed dimension $d\ge2$. In particular, for $d>4$, the population risk of our modified problem can be shown to have equivalent convergence rates as the original least square convex regression estimator of $\widetilde O(n^{-2/d})$ using neighborhoods sizes (i.e. $K=K_n$) of order $\widetilde O(n^{2(d-1)/(2d-1)})$ for general convex functions and $\widetilde O(n^{(d+1)/(d+2)})$ for strongly convex functions. To corroborate our theory, we implement iterative optimization methods to compare the prediction accuracy of the KNN relaxation with those of the full estimator across synthetic convex benchmarks. Our experiments demonstrate that KNN prediction risk approaches the full-estimator prediction risk at neighborhood sizes matching the theoretical scale.

0.35

Scalar Smoothness Is One Point on a Besov Spectrum

cs.LG cs.CV ·2026-11-01 ·Bum Jun Kim, Anastasis Kratsios
Abstract · 83 pages, 10 figures

Studies of neural network generalization and robustness often summarize smoothness with a single number, such as a Lipschitz constant or a Hessian curvature statistic. However, Lipschitz constants summarize global worst-case sensitivity and Hessian statistics summarize local curvature; neither captures how model responses vary with the size, direction, and location of perturbations encountered in real-world deployment. To measure smoothness under these perturbations, we introduce a finite-difference roughness spectrum that profiles model responses across these conditions. Our roughness spectrum generalizes scalar smoothness measures, bringing Lipschitz sensitivity and directional Hessian curvature into a common framework across perturbation scales. We connect this spectrum to classical Besov theory, which describes how a function's variation changes with scale. Our theoretical analysis also proves that a larger output shift can cause less task damage than a smaller one, and that local curvature alone can miss the blocks most vulnerable to finite perturbations. In image classification, Roughness Source Control (RSC) suppresses margin roughness along noise directions and improves noise robustness in a comparison of 80 checkpoints across four model--dataset combinations. In post-training quantization, we identify blocks that need higher precision by measuring task-loss increases from perturbations matched to the size and direction of each block's quantization error. Our mixed-precision allocation achieves lower perplexity than the Hessian-based HAWQ-V2 baseline under the same GPTQ weights and bit budgets in all 14 settings across seven held-out large language models.

0.34

Complete Donor-Set Guidance for Frozen Diffusion Docking Models

cs.LG physics.chem-ph ·2026-11-01 ·Hyunwoo Kim, Insoo Park, Ziseok Lee, Wooyeol Lee, Yoorim Gang +1 more
Abstract · 38 pages including appendices, 4 figures. Hyunwoo Kim and Insoo Park contributed equally

Metalloprotein docking requires a ligand pose to satisfy both global pocket fit and the identity and geometry of metal-coordinating atoms. General-purpose diffusion docking models do not explicitly condition generation on externally specified complete ligand donor sets, and their predicted poses can exhibit incorrect metal-donor relationships. Metal-aware models address coordination during training and generation; here we ask whether external donor information can correct already-trained general-purpose models without retraining. We propose Complete-Identity Contract Guidance (CICG), which applies a supplied complete donor set through translation, rotation, and torsion updates, and Historical Contract Transfer (HCT), which supplies donor identities from earlier complexes at compatible receptor sites. We compare HCT with trajectory-local controls retaining the historical donor count with or without elemental composition. On 86 assessable HCT-covered queries, HCT-guided CICG increases frozen-confidence-selected joint success from 21.5% to 36.6% on SurfDock and from 25.0% to 34.9% on DiffDock-L. The joint criterion requires correct global ligand placement, complete native donor recovery, coordination geometry, preserved stereochemistry, and absence of fatal receptor clashes. Composition-matched guidance reaches 33.14% and 34.30%, leaving the additional benefit of fixed historical membership unresolved. Our results therefore support donor-set guidance as a conditional correction for frozen docking models, while identifying donor-source reliability and physical-quality preservation as the main barriers to broader use.

0.34

Robust Remote ECG Monitoring under Energy Constraints: Anomaly Filtering versus Predictive Uncertainty

cs.LG cs.PF ·2026-11-01 ·Mustafa Fuad Rifet Ibrahim, Maurice Meijer, Alexander Schlaefer, Peer Stelldinger
hot direction
Abstract · 21 pages, 7 figures

Machine learning for remote electrocardiogram monitoring must handle noise and diseases absent from training under tight resource constraints. We investigate whether upstream unsupervised anomaly detection can improve monitoring decisions more efficiently than classifier-derived uncertainty. We evaluate five detectors on PTB-XL and BUT QDB and compare a two-stage filter-classifier system with Monte Carlo Dropout and deep ensembles under unseen-disease and noise conditions. Performance is averaged over five training seeds. We measure single-pass energy, latency, and graph memory on a microcontroller's central processing unit (CPU) and neural processing unit (NPU). Per-score neural network costs account for repeated evaluations but exclude scoring outside deployed graphs. Energy equations relate filtering and rejection to avoided classifier execution. Across four disease scenarios, Deep Support Vector Data Description (Deep SVDD) achieves the highest mean area under the receiver operating characteristic curve within all tested finite CPU energy budgets. On the NPU, it leads at 1 mJ, while Monte Carlo Dropout has slightly higher means at 10 and 100 mJ. With a narrow CPU classifier, the Deep SVDD two-stage system increases the fraction of correctly classified or correctly rejected inputs by 15.0-19.4 percentage points over classification without rejection. Estimated energy per input is 1.93-6.99 mJ, compared with 52.53 mJ for Monte Carlo Dropout and 8.76-17.51 mJ for deep ensembles, at each method's retrospectively selected test accuracy peak. These results support upstream filtering for efficient remote monitoring, while independent validation and complete-system measurements remain necessary.

0.34

Nonreciprocal Spherical Attention and Retrieval: Sharp Coarsening Clocks and Certified Score Correction

cs.LG cs.IRmath.DS ·2026-11-01 ·Faruk Alpay, Taylan Alpay
benchmarktheory
Abstract · 42 pages, 9 figures, 10 tables. Includes mathematical appendices and ancillary reproducibility code

We develop nonreciprocal spherical attention as a mathematical framework for coarsening and compressed nearest-neighbor retrieval. For two shrinking acute blocks whose centers have zero skew pairing and whose total mass ratio is subexponential in inverse temperature, we prove $\lim_{β\to\infty}β^{-1}\log T_β=1-\cosφ$, without restricting particle-count growth. An explicit three-body flow has two rigorously separated metastable exponents. For finite compression, mass centering yields a sharp $O(r^2)$ attention-output error bound at fixed temperature, with an explicit spherical-normalization correction. We formulate skew query-side score calibration as a strictly convex moment problem and derive a unique normal equation, residual-based optimization certificates, and bounded low-rank projections. A norm-aware margin objective calibrates native squared-distance rankings. Experiments use semantic retrieval data, one million visual descriptors with three codebook seeds, and a ten-million-vector FAISS IVF-PQ index with recomputed exact cosine ground truth. On SIFT1M, skew margin calibration raises native-L2 hit@10 from 0.8926 to 0.8948; symmetric calibration reaches 0.8996. The ten-million-vector index reaches 0.8827 hit@10 at 0.1794 ms per query in a four-thread batch benchmark, while validation selects no skew update. Exact streamed attention over ten million keys evaluates compressed memory on 1,024 held-out queries. Independent matrix solves, automatic differentiation, finite differences, interval arithmetic, full-flow tests, and nine Rocq-checked algebra lemmas support reproducibility. Directionality alone does not establish retrieval superiority.

0.34

SPAR: Experience-Driven Strategy Evolution in Multi-Session B2B Sales

cs.AI ·2026-11-01 ·Hangming Zhang, Yao Zhao, Haotian Kuang, C. L. Philip Chen, Tong Zhang +1 more
top venue
Abstract · EMNLP 2026 Industry Track

B2B outbound sales involves high transaction values and relationship-based trust, so closing a deal typically requires multiple rounds of outreach building on prior context. Existing LLM-based sales agents operate within single sessions and cannot carry forward context or refine strategies from experience. Methods that do learn from experience store lessons as unstructured text and treat all outcomes as equally informative, with no mechanism to distinguish a flawed strategy from an external factor. We propose SPAR, a self-evolving framework that models selling strategies as testable hypotheses and refines them without parameter updates. A double-loop architecture lets an inner loop adapt to individual merchants through episodic memory while an outer loop evolves a shared strategy playbook through causal attribution and hypothesis-driven evolution. In a live A/B test on the 1688 platform with over 200,000 merchants, SPAR achieves +15.8% conversion lift and +35.8% high-intent merchant growth. Human evaluation by domain experts rates SPAR at 4.62 out of 5 on overall sales quality, the highest among all methods.

0.34

ECGJev: Question-Conditioned Option Scoring for 12-Lead ECG Interpretation

cs.LG eess.SP ·2026-11-01 ·Zhuoyuan Li, Xidong Wu, Hanyu Cui, Jiansong Gao, Rongrong Fu +2 more
code/datahot direction
Abstract · 5 pages (4 pages plus ethics statement and references), 2 figures. Submitted to IEEE ISBI 2027. Zhuoyuan Li and Xidong Wu contributed equall…

Clinical interpretation of a 12-lead electrocardiogram (ECG) often requires answering targeted diagnostic questions with clear, quantifiable decisions. We introduce ECGJev, the first ECG framework built on a decision decoder, a pretrained language model trained to return a probability for every listed option. A question-conditioned bridge compresses frozen waveform and spectrogram features into 16 ECG tokens, and a two-stage procedure adapts the bridge and low-rank adapters on the decoder's native option readout using diagnostic labels alone, scoring all options for a question in one forward pass. Evaluated across four ECG benchmarks, ECGJev achieves the highest macro-F1 on CPSC2018 (80.46%), PTB-XL (71.68%) and Chapman (96.86%), and generalizes competitively to the small-sample PTB cohort. Concept probes and perturbation analyses further show that its predictions rely directly on the ECG input, remain stable under option-order and wording variations, and transfer zero-shot to unseen waveform concepts. Our code is available at https://github.com/zhuoyuanli-lab/ECGJev.

0.34

CoGSeq-LTV: Scalable Missing-Aware Graph-Sequence Modeling for Multi-Business Transaction Analytics

cs.AI ·2026-11-01 ·Ziyang Su, Yao Zhou, Kelin Liu, Guanjie Zheng, Linghe Kong
Abstract · submitted to WWW2027

Large-scale super-apps generate heterogeneous transaction logs in which users interact with multiple businesses, item identifiers are only partially available, and future value targets are sparse, heavy-tailed, and cumulative over time. We study multi-business, multi-horizon value prediction as a data-intensive analytics problem over such logs, rather than as a single-table task. (i) a batch-induced user-item graph that provides collaborative signals without materializing a global graph and remains applicable when item identifiers are missing through reserved unknown-item handling, (ii) business-aware sequence encoders that preserve recency, inter-event gaps, payment context, and goods-table availability, and (iii) cross-view contrastive alignment that transfers user-level intent across businesses while retaining business-specific temporal patterns. To handle zero-inflated and long-tailed cumulative targets, CoGSeq-LTV adopts a zero-inflated mixture log-normal head with MAE-aligned point supervision and validation-calibrated zero-aware inference. Experiments on the BiBiSheng transaction-log dataset show that CoGSeq-LTV improves over representative LTV baselines across businesses and horizons, including under a later out-of-time evaluation window.

0.33

Auditing a Model-Hub Benchmark: Data Quality, Training Budgets, and Side Information in HuggingBench

cs.LG ·2026-11-01 ·Michal Podstawski
benchmark
Abstract · 12 pages, 10 tables, 2 figures. Code and data: https://doi.org/10.5281/zenodo.23857277

Benchmarks built from model hubs combine interaction data, knowledge graphs (KGs) and resource documentation, but valid files do not ensure that these artifacts describe the same objects. Treating the release as a data product, we audit HuggingBench (SIGIR 2025), a benchmark for recommending Hugging Face resources, through data validation, a replication of its eleven recommendation methods and controlled experiments with three training seeds. In three of its four external KGs, the files pass key and range checks but violate the item-entity identity constraint on which KG recommenders rely. Correcting this join improves every KG-based method in every seed, by up to 73% in mean Recall@10, and consistently reverses ten of 27 comparisons between KG configurations. The protocol matters as well: the 100-epoch budget ends before early stopping in seven to eight of 29 validation-selected settings, and 21 of 23 such setting-seed pairs improve with a longer budget. The released item text, which no reported method uses, is the strongest signal: closed-form linear models with text and the item graph exceed the best reported Recall@10 by 18% while fitting in minutes on a CPU. Checks of other recommendation releases show that individually valid releases of one dataset can use incompatible identifiers. We release corrected files, mappings, validators and logs, and derive practices for preserving identity and provenance across the artifacts of benchmarks and model lakes.

0.33

Counterexample-driven autonomous research with application to excess radiated energy in a Z-pinch

cs.AI physics.plasm-ph ·2026-11-01 ·Bowen Zhu, Jian Wu
code/data
Abstract · 27 pages, 8 figures, 5 tables; v2 adds fine-grid Stage IV trailing-plasma timing and timestep-contraction controls, load-versus-background e…

Recent breakthroughs in artificial-intelligence-assisted mathematics show that open-ended machine search can outperform human mathematicians, yet transferring this paradigm to physics requires refactoring the research workflow. We introduce Simjecture, an open-source framework in which language-model agents pursue human-defined physical conjectures. The framework tackles two central challenges in this transfer: first, because physics is empirical and no model is final, it structures inquiry around counterexample-driven falsification and prospective hypothesis repair; second, because physical theories must interact with the real world, it uses computational physics as an executable testbed while separating the qualification of numerical instruments from scientific claim review so solvers and diagnostics can be adapted within a recorded investigation. We apply the framework in a 136.3 h investigation to tackle the long-standing problem of radiated energy exceeding the imploding-shell kinetic energy in a Z-pinch. The investigation reveals competing effects of energy conversion, retention, and release timing. Stronger coherent helical deformation increases resistive conversion while reducing radiation and increasing retained energy. In a matched fine-grid test, broadening the trailing-plasma profile delays the later radiation maximum by 1.80 ns and recovers an early 153 J bolometric deficit to within 0.08% by 44 ns, with a spectral response that shifts during the pulse. These results provide new insight into energy conversion in high-energy-density plasma physics and demonstrate a new paradigm for physics research.

0.33

A Compute Center Anyone Can Join: A Peer-to-Peer System for AI Inference

cs.AI cs.CRcs.DCcs.GT ·2026-11-01 ·Yimo Lin, Zhen Zhang
Abstract · 13 pages, 2 figures, 1 table. Ancillary files: model script reproducing all numerical results, and black-box probe data

Today, most people obtain AI inference by buying it from a small number of organizations that run large data centers, while many GPUs owned by individuals and small institutions sit idle for much of the day. Bitcoin showed that strangers can contribute resources under shared rules and receive checkable rewards, but it relies on an asymmetry: proof-of-work is expensive to produce and cheap to verify. Useful inference has no such asymmetry. We propose the Open Inference Market (OIM), which replaces the missing asymmetry with an economic mechanism. A worker delivers a result without knowing whether it will be audited; its unsettled earnings serve as collateral during a challenge window; and confirmed cheating forfeits all of it. We call this principle earnings as collateral. In a simplified model we show three things. The number of audits needed for deterrence is a constant per worker per day, independent of the worker's throughput, so the audited share falls as the network gets busier. A new worker can join with a short period of intensive auditing and a very small bond. And whether the mechanism works at all depends only on the ratio of the audit judge's false-flag rate to its detection rate. Under example parameters, about 0.3 replays per worker per day are enough to make cheating unprofitable for a rational worker. On this basis we lay out a path that hands four powers back to participants one at a time: owning compute, writing history, judging work, and releasing value. A compute center need not be a building; it can be a protocol that anyone can join.

0.33

Who Asks, Who Answers: Linear Attention Builds the Query in Hybrid Language Models

cs.AI ·2026-11-01 ·Wenqi Zeng, Chen Feng, YuAng Chen
code/data
Abstract · 10 pages, 4 figures

Hybrid language models interleave linear-attention layers with a small number of softmax-attention layers. Prior work largely attributes in-context retrieval to softmax attention, treating linear attention primarily as a mechanism for compressing past context. We revisit this division of labor and show that linear attention plays a direct role in retrieval by constructing the query used by softmax attention. Studying Qwen3.5 (Gated DeltaNet) and Granite4.0 (Mamba-2), we identify an Encode--Address circuit. An early Encode block preserves token identity, while an Address block constructs the final-position representation from which a softmax \emph{reader} forms its attention query. Transplanting this representation between prompts with different queried keys redirects retrieval in 0.97--0.99 of cases. The Address block's recurrent state carries the retrieval target, and its computation transfers across architectures: replacing Granite's Address block with Qwen's restores retrieval accuracy to 0.92--0.94 without updating either model's weights. Across six Qwen3.5 models, Address computation becomes distributed at larger scales over a chain of blocks, each guiding a successive reader, as the query no longer passes from one reader to the next through attention. These findings reveal a division of labor in hybrid retrieval: linear attention constructs the query, while softmax attention performs the lookup. Code is available at https://github.com/yuang-chen/LinearQuery

0.33

BIOTRAM: Biochemistry-informed Transformer for Relative Abundance via Metabarcoding

cs.LG q-bio.GN ·2026-11-01 ·Emma Boehly, Scott C. Lowe, Ken A. Thompson, Anthony Fuller, Will Armstrong +68 more
Abstract

Arthropod monitoring at scale begins with metabarcoding of bulk samples: trap collections of hundreds to thousands of specimens processed as one unit, extraction of DNA from the bulked specimens, then sequencing of the DNA from all specimens together. Metabarcoding reveals which taxa such a sample contains, but due to factors such as differing DNA amplification rates and body sizes, its read counts do not neatly correspond to how abundant the different taxa are relative to one another. This lack of relative abundance limits the data's utility for the multitude of ecological analyses that depend on it. We propose BIOTRAM: a neural network that estimates per-taxon specimen counts by dividing the observed read counts by a learned read yield per specimen, which it factors into the DNA each specimen contributes and how efficiently that DNA is amplified. On a paired dataset of 1,188 bottles (800 k specimens) processed both by metabarcoding and by comprehensive individual barcoding for ground-truthing, BIOTRAM reduces the Jensen-Shannon divergence between the estimated and the true composition from 0.3586 for raw read proportions (the usual proxy for abundance) to 0.1200. Trained on the same objective and inputs as four generic deep architectures, all with tuned loss weights, it reaches a count error 3.2% below that of the strongest of them and the lowest composition divergence. It also generalizes to out-of-domain data better, beating all baselines on bottles from unseen regions, later collection dates and unseen sequencing batches, and transferring to data from an unseen laboratory. Structuring the mapping from reads to counts around the steps of the experimental pipeline thus yields more accurate abundance estimates than leaving that mapping unconstrained.

0.33

Reduced Cost Influence Functions for Predict-then-Optimize under Noisy Data

cs.LG stat.ML ·2026-11-01 ·William Zhang, Saurabh Amin, Georgia Perakis
Abstract · 10 pages, neurips submission

Machine learning models deployed in decision-making pipelines, from vehicle routing to dynamic pricing to clinical decision support, must remain reliable in the face of low-quality training data. We study a predict-then-optimize setting in which an upstream model predicts the cost vector of a downstream linear program, with a large pool of noisy training data supplemented by a small curated set of clean data. We introduce Reduced Cost Influence Functions (RCIFs), which combine influence functions from robust statistics with the geometry of linear optimization to estimate how each training point shifts downstream decisions. Building on this, we propose a reweighting algorithm that uses RCIFs to downweight harmful training points and reduce downstream regret. Experiments on random linear programs and shortest-path problems show that our approach reduces regret, and accompanying theory characterizes properties of RCIFs and identifies regimes favorable to our method relative to baselines.

0.32

Generative Modeling with Continuous Flows: Sample Complexity of Flow Matching

cs.LG cs.AI ·2026-11-01 ·Mudit Gaur, Prashant Trivedi, Shuchin Aeron, Amrit Singh Bedi, George K. Atia +1 more
theoryhighly cited
Abstract

Flow matching has emerged as a promising alternative to diffusion-based generative modeling, offering a continuous-time ordinary differential equation formulation with fast sampling and a simple regression-based training objective. Despite its empirical success, some theoretical analyses idealize the learning procedure by assuming that the velocity field is obtained by exact empirical risk minimization. This assumption does not reflect the practical finite-dataset setting, where the empirical flow-matching loss is only approximately minimized by a finite number of stochastic-gradient updates. We develop an empirical-risk-minimizer-free analysis for flow matching that explicitly accounts for both finite-sample statistical error and finite-time empirical optimization error. Our analysis considers a finite dataset of \(N\) flow-matching samples and a learned parameter \(θ_T\) obtained after \(T\) stochastic gradient updates on the empirical objective. Under regularity assumptions on the interpolation path, sub-exponential source and target distributions, approximation by a bounded-parameter smooth neural velocity class, empirical-loss optimization, and uniform flow stability, we prove a guarantee that holds with high probability over the finite training dataset and in expectation over the stochastic-gradient randomness. The resulting bound separates approximation error, statistical error from the finite dataset, tail error from unbounded source and target distributions, and empirical optimization error from approximate minimization of the finite-sample objective.

0.32

ParetoSaddler: Pareto-Guided Optimization of Agent Harnesses for Quality and Cost

cs.AI cs.CLcs.LGcs.MA ·2026-11-01 ·Xingzhi Qian, Sungho Park, Rongyuan Tan, Jianbo Yu, Jue Zhang +11 more
community pick
Abstract · Project website and code will be available at https://aka.ms/ParetoSaddler-website

Harness optimization can substantially improve LLM agents without modifying model weights, but existing methods primarily optimize response quality. Extending them to joint quality--cost optimization introduces a challenge: applying the deployment preference during intermediate updates can prematurely reject candidates that are valuable for later improvements. We introduce ParetoSaddler, which addresses this mismatch by separating intermediate candidate promotion from final deployment selection. At each iteration, an optimizer updates a parent harness to produce a candidate, and then compares the two over repeated executions and uses Bayesian bootstrap Pareto support to promote candidates showing credible progress in either quality or cost. Promoted candidates receive broader development evaluation and remain available as parents for later optimization steps. We evaluate ParetoSaddler on ClawMark and GAIA2 using domain-disjoint splits. On ClawMark, ParetoSaddler improves quality from 70.86\% to 81.35\% while reducing cost from \$1.17 to \$1.01 per task; on GAIA2, it improves quality from 56.22\% to 62.89\% while reducing cost from \$0.78 to \$0.45 per task. The Pareto-selected harnesses also achieve higher quality at lower cost than quality-only harness optimization. Analysis shows that Pareto-aware optimization retains intermediate candidates that would be discarded by deployment preference but can support later improvements, while Bayesian bootstrap screening filters out candidates whose gains are weakly supported across runs.

0.32

LLM-Enhanced Multimodal House Price Prediction using Structured, Textual, and Visual Features

cs.LG cs.CLcs.CVcs.IR ·2026-11-01 ·Khandokar Md. Rahat Hossain, Souvik Das, Tahsina Hashem, Mohammed Eunus Ali, Tanzima Hashem
code/data
Abstract · 35 pages, 7 figures, 5 tables. Code: https://github.com/rh2975/LLM-VLM-HousePrice

Accurate house price prediction is essential for real estate buyers, sellers, agents, and financial institutions. Traditional methods rely primarily on structured data, e.g., number of bedrooms, bathrooms, parkings, but they overlook rich contextual information embedded in textual descriptions and property photos. Previous deep learning approaches for house price prediction have fused multimodal data but they fall short to extract structured spatio-temporal and contextual features from texts and semantic visual information from images. In this paper, we introduce a novel multimodal framework that addresses these limitations by systematically integrating structured attributes, textual descriptions, and visual images of properties to achieve higher prediction accuracy while maintaining interpretability. We use a large language model to derive domain-specific spatio-temporal and contextual features from text, and an unsupervised vision-language approach to automatically group photos into meaningful classes like indoor, outdoor, mixed, or floorplan, that are commonly present in real-world house listings. These categories are created without manual labeling by aligning image clusters with carefully designed pseudo text prompts. We fuse all features into a unified representation and use an optimized regressor to predict prices. Experiments on a large real-world property dataset demonstrate that our method outperforms traditional methods and existing multimodal baselines. Ablation studies confirm that the novel textual feature extraction and categorized visual embeddings substantially contribute to more accurate and reliable price predictions.

0.32

Opening the Black Box of Neural Auction Mechanisms

cs.AI ·2026-11-01 ·Li Kong, Zilong Liu, Haoyang Zhou, Hanbing Liu, Qi Qi
code/data
Abstract

Neural auction design has emerged as a powerful paradigm within automated mechanism design for modern Web platforms. Because these learned mechanisms directly dictate critical economic outcomes, their lack of internal transparency raises significant auditability and trust concerns. Opening this black box \textit{post-hoc} is therefore essential, yet presents two distinctive challenges: (1) activation contexts are dense, correlated numerical profiles devoid of explicit semantic structure, and (2) feature interpretations can only be falsifiably audited if they map to measurable auction states or outcomes. In this paper, we propose a post-hoc white-box auditing framework for unpacking trained neural auctions. Using CITransNet as a representative testbed, we first employ sparse autoencoders to decompose hidden representations while faithfully preserving downstream auction behavior. Anchoring feature interpretations to native economic targets, we validate them through activation prediction and causal interventions to isolate verified features from the reconstruction dictionary. Finally, applying this pipeline across varied auction settings, we show that the verified features can serve as analytical instruments to examine how internal representations modulate strategic misreporting incentives, align with Myerson's optimal auction, and support the diagnosis of individual auctions. Our work highlights the potential of white-box mechanism auditing to build the foundational trust necessary for neural auctions to be deployed as widely adopted Web technologies. To enhance research transparency and facilitate independent inspection, we publicly release the experimental results supporting our findings at https://github.com/Li-Beverly-Kong/opening-neural-auctions.git

0.32

Toward Agentic AI-Ready Workflows in Built Environment Organizations: Knowledge Structures and Interoperability Challenges

cs.AI ·2026-11-01 ·Ahmet Bahaddin Ersoz, Justin McHenry, Yevgeniya Kovalchuk, Edward Barrett, Mandana Sarey Khanie
published
Abstract

The rise of agentic artificial intelligence (AI) in the built environment is shifting attention from isolated AI tools toward autonomous systems that reason, plan, and act across project and organizational workflows. The effectiveness of such systems depends not only on large language model capabilities, but also on the readiness of heterogeneous data environments across the Architecture, Engineering, and Construction (AEC) sector. These environments combine structured sources, such as CAD and BIM models, schedules, cost databases, and asset registers, with unstructured ones, including technical reports, design narratives, planning and policy documents, contracts, meeting minutes, and risk registers. This paper reviews how built environment data can be organized to support agentic AI-ready workflows, focusing on knowledge structures, interoperability, and access and integration mechanisms such as retrieval-augmented generation (RAG) and the Model Context Protocol (MCP). Drawing on existing literature, it examines approaches proposed to organize structured and unstructured data and considers how such data can be contextualized and made accessible to AI agents in a reliable manner. The paper synthesizes current perspectives, highlights challenges around fragmentation and interoperability, and identifies open questions for future research on AI-compatible information management in the built environment.

0.31

Metamorphic Testing for Evaluating Robustness of Sentiment Analysis Models

cs.LG cs.AIcs.CL ·2026-11-01 ·Sowjanya Karanam, Jayanth Bhargav
Abstract · 8 pages, 3 figures

Machine learning systems are challenging to test using traditional software testing methods because their behavior is learned from data rather than explicitly programmed. This challenge is particularly significant in natural language processing (NLP), where small input variations can lead to unexpected model behavior. Conventional evaluation metrics, such as accuracy and F1-score, often fail to capture such inconsistencies. Metamorphic testing addresses this limitation by evaluating expected relationships between original and transformed inputs. With the increasing prevalence of generative artificial intelligence (GenAI) and language-based AI systems, the need for systematic robustness testing has become even more pronounced. Motivated by this growing need, this paper presents a lightweight, batch-based metamorphic testing approach for evaluating the robustness of such systems. In particular, we consider sentiment analysis as a representative NLP task and evaluate a pretrained classifier under four transformations: synonym replacement, typo injection, word-order variation, and negation. Using consistency score, flip accuracy, and confidence change, we identify strengths in handling semantic variations and vulnerabilities to textual noise and logical transformations. These findings highlight the value of metamorphic testing in identifying behavioral inconsistencies beyond conventional performance metrics and demonstrate its potential as a systematic framework for evaluating the robustness of language-based AI systems and guiding performance improvements before deployment.

0.31

MemHarmBench: Evaluating Cross-Session Prompt Injection in Memory-Persistent AI Agents

cs.AI ·2026-11-01 ·Zhihua Wang, Yunhao Feng, Yifeng Gao, Rui Pan, Baiheng Zhao +2 more
benchmarkcode/data
Abstract · 24 pages, including appendix

AI agents often rely on persistent memory to retain user preferences and operational rules across sessions. However, this capability introduces a new safety risk: malicious instructions from messages, documents, or Web content can be retained in memory and influence future tasks even after the original attack content is removed. Existing attack success metrics, however, fail to capture who is affected and how attacks translate into downstream harm during otherwise normal task execution. We introduce MemHarmBench, a victim-centric benchmark for evaluating cross-session prompt injection in memory-persistent AI agents. MemHarmBench considers three categories of potential victims: users, workspace/project owners, and external stakeholders. Our evaluation first exposes an agent to malicious instructions and tests whether they are retained in memory. We then remove the attack content and initiate a new session with a legitimate task while preserving the agent's memory. We jointly evaluate whether the agent follows the attacker's intent, completes the legitimate task, and exhibits observable execution anomalies. Across multiple agent systems, we find that cross-session attacks can remain behaviorally stealthy: agents often complete tasks normally and exhibit no detectable anomalies while still making critical decisions aligned with the attacker's intent. By connecting malicious memory retention to downstream decisions and affected parties, MemHarmBench provides a systematic framework for exposing safety risks that conventional attack- and task-success metrics may overlook. We publicly release MemHarmBench at https://github.com/Kirito14IT/MemHarmBench.

0.31

Decoupled Predictive Decision Making: A Shared Conditional Predictor for Agents

cs.AI ·2026-11-01 ·Lang Mo, Xingxuan Zhang, Qi Wang, Mingchao Hao, Hao Yuan +3 more
Abstract · 16 pages, 7 figures, 11 tables. Includes appendices

Agent decisions depend on empirical relationships that deployment observations reveal and task descriptions alone may leave unresolved. We introduce Decoupled Predictive Decision Making (DPDM), an architecture that formulates heterogeneous agent decisions as conditional prediction with a shared pretrained learner. Language models provide semantic representations of current and historical records, while a general-purpose conditional predictor predicts task-defined targets from the represented evidence. Pretraining supplies reusable inference capability; deployment evidence instantiates problem-specific predictions without updating either model's neural parameters. Tool judgments, candidate preferences, and reasoning assessments share an evidence-and-query interface, with predictions mapped to choices by a decision rule. This architecture establishes complementary sources of agent capability: semantic knowledge acquired through language pretraining and empirical inference acquired through structured-data pretraining. A single frozen LimiX-2 predictor serves all five task categories. With Qwen3.5-27B as the semantic encoder, DPDM raises mean accuracy across 50 benchmarks from 78.80% with few-shot prompting to 84.92% under their respective evaluated history budgets, and reaches 82.33% with identical histories and queries. On ToolPRMBench's BFCL and ToolSandbox tasks, absolute accuracy gains are 34.26% and 28.87%, respectively, at the methods' respective evaluated history budgets. Accuracy improves as history grows. With identical imbalanced histories, mean balanced accuracy improves by an absolute 14.51% across five reasoning-step assessment tasks and four encoders. DPDM makes learning from experience an independently improvable foundation-model capability, creating a shared route for advances in conditional prediction to improve decisions across agent tasks.

0.31

SensOFormer: Foundation Models for Particle Accelerators

cs.LG physics.acc-ph ·2026-11-01 ·Mahindra Rautela, Alexander Scheinker
code/data
Abstract

Large-scale scientific facilities, such as particle accelerators, are complex systems composed of thousands of interacting components that must operate collectively to support diagnostics and control, and ensure safe, reliable operation. These facilities rely on extensive sensor networks that generate large volumes of heterogeneous, incomplete, and noisy telemetry, often sampled at different rates across subsystems. In this paper, we adopt the foundation-model paradigm to learn general-purpose sensor representations from historical accelerator data through self-supervised pretraining. We introduce SensOFormer, a denoising masked transformer with a Perceiver-style encoder-decoder architecture designed to handle variable sensor sets, accommodate heterogeneous sampling rates, and learn robust representations from noisy measurements. The model is pretrained on multiple particle-accelerator datasets and transferred to downstream tasks, including missing-data imputation, anomaly detection, cavity identification, and fault identification. Through extensive evaluations and comparisons, we demonstrate that a single self-supervised model can learn reusable representations for diverse operational tasks in large-scale scientific facilities. Project page: https://mahindrautela.github.io/sensoformer.

0.30

Correctness and Predictive Uncertainty in Bayesian TSK Fuzzy Regression: A Reproducible Evaluation on Tabular and Building-Meter Data

cs.LG stat.ML ·2026-11-01 ·Zhiren Xiao
unusual
Abstract · 21 pages, 3 figures. Supplementary material is included as an ancillary PDF

Uncertainty estimates from fuzzy regression are informative only when the implementation is correct and the comparison protocol makes the sources of variation visible. We revisit a first-order Takagi--Sugeno--Kang (TSK) regression implementation and rebuild its dense, conjugate Bayesian, BIC-based, and rule-level spike-and-slab estimators using corrected code. The corrected pipeline freezes the Gaussian membership partition at training time, uses a scaled conjugate prior, evaluates firing strengths in the log domain, and reports direct posterior-predictive quantiles for Bayesian methods. Across 30 outer 80/20 splits on three targets from two UCI regression datasets, corrected TSK-LS obtains $R^2=0.957$, $0.923$, and $0.786$ on Energy-Heating, Energy-Cooling, and Concrete, respectively; rule-level Gibbs sampling with Bayesian model averaging obtains $0.957$, $0.922$, and $0.784$. We then evaluate the same workflow on a bounded chronological sample of the Building Data Genome 2 meter data that covers 158 buildings across 17 sites. In a common-unit evaluation, every method is scored on the same 47,400 test rows and every trained model uses 94,800 train--validation rows. Persistence has RMSE $27.710$, compared with $29.083$ for both TSK-LS and Bayesian TSK; the pre-specified building-level mean RMSE difference between Bayesian TSK and persistence is $+0.0447$ with a 95% site-cluster bootstrap interval of $[-1.5536,1.5665]$. The Bayesian TSK 90% intervals have PICP $0.980$ and MPIW $116.469$. Taken together, the evidence supports a narrower conclusion: spike-and-slab model averaging is a reproducible implementation option, but it does not provide a demonstrated accuracy or calibration advantage across the tested tabular and building-meter settings.

0.30

ZeroGraph: Compositional Blockwise Training via Materialized Boundary Supervision

cs.AI ·2026-11-01 ·Omar Ramadan, Sam Siavoshian, Amir Kashif Saeed, Benjamin A. Johnson, Amin M. E. -A. Diab +1 more
code/data
Abstract · 15 pages, 5 figures. Code available at https://github.com/oramada/ZeroGraph

Blockwise learning can remove online gradient dependencies across depth, but independently optimized blocks need not compose into a useful model. We present ZeroGraph, a systems transformation for pretrained-model conversion and distillation that materializes a frozen teacher's coherent boundary trajectory as a versioned dataset. Each partition then trains with disjoint parameters, optimizer state, random state, and checkpoints, replacing live cross-depth dependencies with persistent supervision. Experiments separate learning quality from systems behavior. On GPT-2 Large, an unanchored objective reduces denoising error by 99.44%, yet assembled held-out cross-entropy worsens from 3.63 to 10.48 and context use becomes negative. A six-arm TinyLlama study identifies clean teacher-boundary regression as the successful interface; the tested noisy objective fails with cached or online targets. The clean recipe passes quality and context-use gates across three Qwen2.5-1.5B seeds and on a 3.086B Qwen2.5 model. In an objective-matched comparison, cached and online targets are bitwise identical. Cached jobs are faster after materialization, but a 618.31 s cache-build cost prevents wall-time break-even through 16 measured reuses. Boundary-shift tests show tolerance to intermediate representation mismatch, and an interrupted block resumes to a bitwise-identical final state without rerunning unaffected blocks. In the tested pretrained-model conversion and distillation setting, cross-depth semantic dependence can move from online optimization into persistent data.

0.30

Recursive Reasoning Exploration: Mathematical Research through Evolving Reasoning Graphs : case study: An Exact Certificate-Degree Theorem for Two-Term Tensor Decompositions : Phase III of the Math Instruments Program

cs.AI math.HO ·2026-11-01 ·Ying Xie
publishedtheory
Abstract

Large collections of manuscripts and research notes offer ways to reuse arguments, while making it harder to retain hypotheses, check claims, and understand how results were obtained. We propose Recursive Reasoning Exploration (RRE), a framework for cumulative mathematical research through evolving reasoning graphs. RRE links claims and jointly required premises to research targets, attempted arguments, counterexamples, checks, and revisions. Each investigation can generate further targets, and its supported outcomes become available to subsequent investigations. This is Phase III of the Math Instruments program: identifying mathematical instruments, reconstructing reusable reasoning, and organizing recursive research from that reasoning. The framework supports human-directed, collaborative, and goal-directed autonomous exploration, with open-ended exploration as a further mode. Its records also provide material for learning how mathematical investigations develop. A complexity research case study establishes an exact theorem: for every complex tensor of order n >= 2 and rank greater than two, the minimum Nullstellensatz certificate degree for its original two-term decomposition equations is 3n. This includes an infinite family whose every flattening has rank at most two. The exploration develops an explicit boundary certificate, a balanced transfer lemma, and an exhaustive argument using classical rank-two geometry. A parallel branch gives bounds 3m + 3 <= delta_m(I_m, N) <= 3m + 6 for every nonzero nilpotent m by m matrix. Six graph revisions record the progression from inherited results to these theorems, with exact arithmetic checks and a replayable exploration history. We present the RRE representation and protocol, the mathematical proofs, visualizations of graph evolution, and a research agenda for shared human-AI exploration, cumulative reuse, and education.

0.30

CREDO: Evidence-Conditioned Credit Assignment for Temporal Reasoning Agents

cs.LG cs.AIcs.CLcs.IR ·2026-11-01 ·Yue Leng, Shengtao Wen, Xiang Chen, Xinyi Zeng, Yu Tian +1 more
code/data
Abstract · Code available at https://github.com/NUAA-MMMI/CREDO

Temporal Knowledge Graph Question Answering (TKGQA) requires language model agents to retrieve evidence, select relevant facts, and perform multi-step temporal reasoning. However, the trajectory-level outcome signal in GRPO does not explicitly distinguish the quality of intermediate decisions: downstream failures can penalize useful evidence acquisition, while identical rewards in all-failed groups eliminate the relative outcome signal despite differences in intermediate progress. We propose Credit Redistribution with Evidence-Driven Opportunities (CREDO), which uses intermediate evidence to recover local supervision, including in all-failed groups, and determine where outcome supervision acts. CREDO assigns positive stage-level credit through comparisons among rollouts of the same question and uses retained answer overlap as an opportunity proxy to shift part of the outcome-policy weight toward reasoning and answering. The framework preserves terminal rewards and trajectory-level advantages, while redistribution conserves total outcome-policy weight per trajectory. Following the public evaluation protocol of prior work, CREDO raises Overall accuracy from 78.0% to 82.4% and Multiple accuracy from 55.0% to 65.6% on MultiTQ relative to the strongest published baseline in our comparison. On TimelineKGQA, it raises cross-KG accuracy from 30.9% to 36.4% relative to our reproduced baseline via direct transfer to ICEWS Actor under matched training and evaluation settings.

0.30

Towards a Theory for the Design of Rules in Multi-Agent Systems: the case of Homogeneous Fleets

cs.MA ·2026-11-01 ·Matteo Penlington, Lorenzo Fiaschi, Raj Kiriti Velicheti, Andrea Censi, Emilio Frazzoli
Abstract · 30 pages; A first version of the RDP

The regulation of systems composed of many interacting decentralised agents is a long-standing problem. This problem has typically been resolved by the imposition of rules. Rules resolve the limitations of sub-optimal agents whilst simultaneously aligning the behaviour of the agents with the demands of society. Traffic laws, for example, let a driver cross a junction without seeing around the corner, and transport protocols such as TCP let applications share a network without knowing of one another. The problem is more pressing today, as autonomous systems are deployed in society. In this work, we establish the basis of a system-agnostic theory of rule design for multi-agent systems. We first state the rule design problem (RDP) in its generality. It relates a rule to the set of system performances, defined by the regulator, that the agents deployed under it can attain. However, it is not yet possible to solve the RDP in full generality. As such, we introduce the homogeneous rule design problem (HRDP), in which the agents in each environment are identical, and among other interesting properties, show that under worst-case aggregation the two problems have the same solution, bypassing the combinatorial complexity of the RDP. To solve the RDP under worst-case aggregation we propose Rule Branch and Bound (RuleBnB), an anytime branch-and-bound algorithm. Finally, we illustrate the value of solving the RDP on an autonomous driving case study.

0.30

NorthCUA Technical Report: Turbocharging Computer-Use Agents through Experience at Scale

cs.LG ·2026-11-01 ·Chaoyun Zhang, Bingxuan Li, Pinxue Guo, Yangfan Gao, Zikun Cai +25 more
code/datacommunity pick
Abstract

Scaling computer-use agents from individual actions to complex software work requires experience that connects decisions to complete outcomes. We present NorthCUA, an integrated system for turbocharging computer-use agents through experience at scale. Its Experience Flywheel supplies diverse executable tasks to the NorthCUA Harness, which sustains long interactions, records decisions, and checks task outcomes. The Flywheel uses these records to curate verified trajectories. For unsolved tasks, recursive self-improvement evolves reusable skills and preserves verified milestones so later attempts build on earlier progress. Staged Learning combines this experience with grounding data from the Flywheel to develop GUI competence through mid-training and long-horizon supervised fine-tuning. The policy then learns from fresh attempts in the same Harness through joint desktop-web reinforcement learning, closing the loop between experience production and policy improvement. The resulting Nano, Mini, and Pro family achieves state-of-the-art results among open-source models across GUI grounding, desktop execution, and Web interaction under the reported evaluation settings. For more responsive Web workflows, NorthCUA-Ultrafast pairs direct action selection with open-ended generation on pre-RL Nano, offering a measured quality-latency tradeoff. Together, these advances establish NorthCUA as a high-performing model family and a reusable foundation for agents that carry complex software work from intent to completion.

0.30

NEIGHBOR: Aligning Service Behaviour and Evidence Use in an LLM Agent for Community Resident Services

cs.AI ·2026-11-01 ·Zongyou Yang, Yinghan Hou, Xiaokun Yang
Abstract · 9 pages, 6 figures, 5 tables

Residents who contact a community service desk about a repair or a service charge need answers they can act on, with facts that hold for their case and commitments the desk can keep. In tool-augmented LLM agents, such facts come from documents and records, and the conduct of a service answer is usually specified in a system prompt. We present NEIGHBOR, an LLM agent for community resident services that trains this conduct into the model that drives its tools. NEIGHBOR is deployed on a live resident-service platform for a district of an East Asian city of more than ten million people. It adapts Qwen3-235B-A22B with LoRA on 7,824 supervised fine-tuning examples built from public community-service materials around six service-behaviour targets, among them commitment control and human referral, and the adapted model plans document retrieval and read-only text-to-SQL over the resident's own records. In an offline evaluation with seven configurations, an independent domain expert rated all 350 answers to 50 stratified held-out queries. NEIGHBOR scores 7.74/10 against 5.86 for the same agent running the un-tuned model (+1.88; paired Wilcoxon $p<10^{-8}$; higher on 46 of 50 queries) and improves all six rated dimensions. No tool-augmented answer received a red-line flag (standalone models: 8-14% of answers). An LLM judge ranks the seven configurations much as the expert does (Spearman $ρ=0.93$). Its rubric counts specifics absent from the reference answer as fabricated, so it scores tool-grounded answers 1-2 points lower, enough to change which configuration ranks first. For resident-facing agents, service behaviour is a practical fine-tuning target that raises expert-rated answer quality with the tools held fixed.

0.30

Data-Scarce, Multi-Granular, Multimodal: Benchmark Design for Biosynthetic Gene Clusters

cs.AI ·2026-11-01 ·Hanlin Xiao, Eriko Takano, Mauricio A. Álvarez, Rainer Breitling
top venuebenchmark
Abstract · 43 pages, including references and appendices. Accepted at NeurIPS 2026 (Evaluations and Datasets Track)

Biosynthetic gene clusters (BGCs) are co-located genes that encode the biosynthesis of specialized metabolites, an important source of antibiotics, anticancer agents, immunosuppressants, and agricultural bioactives. Deep learning, particularly self-supervised foundation models, has spurred interest in BGC sequence modeling, but evaluation has lagged behind: existing benchmarks often rely on coarse classification over small curated datasets, providing limited discrimination among models and insight into downstream utility. We propose a benchmark-design framework in which an informative benchmark should be discriminative, separating models with statistical confidence, and applicable, spanning scientifically meaningful tasks under different levels of generalization. We instantiate this framework as BGC-Bench, which derives five tasks from MIBiG, ranging from A-domain substrate prediction to BGC-molecule retrieval, evaluates them under genotype- and phenotype-shift splits, includes an active-learning setting for label-scarce discovery, and is easily extensible to new tasks and encoders. BGCs further provide a shared testbed across granularities and modalities: BGC-Bench compares twelve encoders on nucleotide, amino-acid, or Pfam-domain sequences, including compositional baselines for each input type, and combines BGC and chemical encoders for cross-modal retrieval. No single representation dominates: the best granularity depends on the biological scale of the task, and pretrained encoders do not consistently outperform other representations across tasks and splits. Active-learning experiments show that most tasks remain data-limited, encoder rankings change with labeling budget, and acquisition strategies yield modest gains. These results show that BGC model evaluation requires task-diverse, shift-aware benchmarks, and the framework can be generalized to benchmark design beyond BGCs.

0.30

Black Hole Love Numbers with Physics-Informed Neural Networks

cs.LG gr-qchep-thmath.NA ·2026-11-01 ·Maria J. Rodriguez, Luis Fernando Temoche, Pavlos Protopapas
benchmark
Abstract

We develop and validate a physics-informed neural network (PINN) framework for computing the tidal Love numbers of black holes directly from their perturbation equations. We consider static perturbations of the Kerr black hole, whose exact hypergeometric solutions provide a controlled benchmark for both the numerical solver and the extraction of the physical response. The framework combines exact enforcement of the horizon boundary conditions with an $H^2$-Sobolev loss that incorporates derivatives of the differential-equation residual. The Love number is then obtained by extracting the asymptotic coefficients of the learned solution, without using the analytic connection formula. Across the tested scalar, vector, and tensor perturbations, the method recovers the finite dissipative response with relative errors below $10^{-5}$. Controlled comparisons show that the higher-derivative loss terms improve solution accuracy by orders of magnitude, while independent tests of the extraction procedure distinguish solver errors from errors in recovering the response coefficients. The conservative response, which vanishes analytically, is constrained by the numerical error floor. These results demonstrate that PINNs can accurately compute black-hole tidal response and establish a validated framework for future extensions to frequency-dependent and non-separable perturbation problems.

0.29

Qubrio: High-Performance Quantum Compilation via Multi-Agent LLM Collaboration

cs.AI quant-ph ·2026-11-01 ·Jixuan Ruan, Zhuo Cui, Zhengding Hu, Zhongkai Yu, Xiang Fang +8 more
code/data
Abstract

Large Language Models (LLMs) offer an adaptable alternative to traditional quantum compilation heuristics, which bottleneck progress by demanding costly manual redesigns whenever hardware evolves. However, naive single-pass LLM approaches fail under the unique challenges introduced by quantum compilation. We introduce Qubrio, a practical agentic framework that significantly reduces heuristic redesigns through a three-level decomposition. By (1) partitioning programs into sequential operation stages, (2) deploying specialized agents for placement, routing and optimization, and (3) using disaggregated feedback for precise error correction, our system reliably navigates complex physical constraints. Beyond serving as a direct compiler, our LLM framework also uncovers novel strategies that facilitate convoy-style shuttling, substantially enhancing hardware concurrency without sacrificing fidelity. We further integrate these strategies into the previous state-of-the-art (SOTA) compiler, recovering a significant fraction of the LLM's performance advantages. On realistic workloads, our compiler achieves significant improvements in hardware runtime and program fidelity over the SOTA baseline PowerMove, seamlessly adapting to new hardware capabilities. Qubrio features an open user interface for practical use, and its code is publicly available at https://github.com/crosaa/Qubrio.

0.29

Comparative Strategies for Handling Imbalanced QSAR Regression: A Case Study on hERG Activity Prediction

cs.LG q-bio.QM ·2026-11-01 ·Ming-Yen Sung, Hsin Liu, Tung-Sheng Cho, Cheng-Wei Lee, Yufeng J. Tseng
Abstract

Data imbalance frequently occurs in QSAR modeling, yet most studies focus on classification tasks, and few address regression tasks, as it is challenging to determine which regions of the data are sparse or dense. Data-level and algorithm-level imbalance-handling strategies can have different effects on model performance. In this study, we systematically compare various strategies across GCN and fingerprint-based machine learning models. To evaluate model performance more meaningfully, we partitioned the target region into several subregions based on both domain knowledge and data distribution. We focused on the region of greatest practical interest. Our results demonstrate that selecting models based on this interest-region metric leads to more reliable performance than using the total region metric. We also observed that different imbalance-handling strategies can have divergent effects depending on the model type. These findings highlight the importance of region-aware evaluation in regression QSAR modeling and provide practical guidance for selecting appropriate strategies for imbalanced datasets.

0.29

From Logs to Faults: Reference-Free Fault Localization for Retrieval-Augmented Generation Pipelines

cs.AI cs.SE ·2026-11-01 ·Hira Naveed, Scott Barnett, Stefanus Kurniawan, John Grundy, Chetan Arora +2 more
code/data
Abstract

Retrieval-Augmented Generation (RAG) pipelines are widely used for domain-specific question answering, yet locating faults remains manual and time-consuming. Fault localization is challenging because RAG pipelines consist of interdependent stages, where faults can propagate downstream, masking their origin, or remain undetected even when the final answer is correct. Existing evaluation approaches often focus on individual stages and rely on reference answers or code instrumentation. We present RAGLog, which detects and localizes faults across RAG stages using execution logs, without requiring reference answers or code instrumentation. RAGLog transforms heterogeneous API logs into a common RAG schema and combines rule-based, embedding-based, and LLM-based checks to identify faulty stages. We introduce a live fault injector and RAGLog Faults, a dataset of 3,134 runs from five RAG pipelines with stage-level fault labels. With Claude Haiku 5.5, RAGLog achieves a pooled F1 score of 0.73 on runs with naturally occurring faults, outperforming Claude Code (0.58) and Codex (0.52), and comparable effectiveness on runs with injected faults (0.84 versus 0.80 and 0.82, respectively). It analyzes runs in approximately five seconds at US\$0.005 each, with overall lower cost and latency than agent baselines. RAGLog is available as an open-source tool for debugging, evaluating, and monitoring RAG pipelines.

0.29

METER: Matched-Event Calibration for Short-Event EEG Microstate Transition Analysis

cs.LG ·2026-11-01 ·Abdul Basit, Eklavya Tomar, Muhammad Shafique
Abstract · 36 pages. Supplementary material (16 pages) included as an ancillary file. Abdul Basit and Eklavya Tomar contributed equally

Short cognitive events contain unequal and often sparse transition evidence, so raw EEG microstate distances can reflect finite sampling rather than neural reconfiguration. We introduce METER (Matched-Event Transition Estimation against a Reference), a participant-relative framework that separates marginal occupancy from conditional next-state choice and evaluates each event against duration-conditioned events from a fixed participant-specific reference bank. METER-Jump integrates uncertainty in sparse target and reference transition tables, returning a randomized event rank for testing and a continuous standardized coordinate for association analysis. Using a reproducible three-state representation selected by spatial criteria, we evaluated METER with participant-fitted simulations, estimator ablations, map perturbations, and preprocessing sensitivities. Event-level false-positive rates were 0.044-0.056 across four heterogeneous null mechanisms. By contrast, global cutoffs learned under the participant-fitted null and transferred unchanged to other regimes produced false-positive rates from 0.004 under rare transitions to 0.234 under map perturbation; duration-conditioned ranks remained near 0.05. Detection probability was approximately 0.69 with fixed maps but fell to 0.22 after bootstrap-derived map-assignment perturbation. In 4,175 natural-reading events from 18 participants, no linguistic association survived multiplicity correction in the primary 30-ms analysis, although a 20-ms sensitivity changed one within-stream max-$T$ decision. METER therefore calibrates finite-event departures conditional on a defined microstate sequence, while map assignment, filtering, temporal smoothing, and reference construction remain separate sources of measurement uncertainty.

0.29

ReWY: Faster Backpropagation for Delta-Rule Attention - From 7 to 3 Matrix Multiplications

cs.LG cs.DC ·2026-11-01 ·Yi Zhang, Zhixin Zhao, Weiwei Sun, Yitao Hu, Sheng Chen +1 more
code/data
Abstract

Delta-rule linear attention underlies language models such as Kimi Linear and Qwen3-Next. Chunkwise implementations expose matrix parallelism, but their backward passes compute unnecessary auxiliary adjoints. We present ReWY, an algebraic reformulation that reduces this work before optimizing accelerator execution. Differentiating the chunk correction as one triangular residual system eliminates the inverse adjoint and reduces seven local WY matrix products to three. Two complementary transformations expose further reuse: centered matrix contractions replace pairwise vector-gate processing, and scalar-gate reductions reuse key-gradient products instead of forming a key Gram matrix. These identities simplify backward computation for DeltaNet, GDN, KDA, and GDN2 while preserving the recurrence and gradients in exact arithmetic, without changing the architecture or training objective. We implement the reduced graph on GPUs and Ascend NPUs. Under matched recomputation and full-saving policies, precision-qualified six-shape comparisons with frozen FLA references yield 1.15-1.81$\times$ geometric-mean complete low-level backward speedups on NVIDIA GPUs. All four operators achieve 1.18-1.70$\times$ on H100 and H200 while retaining original FP32 working matrices. A separate public-autograd evaluation on Ascend 950DT, including gradient conversions, yields 1.09-2.89$\times$ under available matched policies. Paired ablations establish further ReWY gains after optimizing the complementary contractions. Complete-gradient checks validate the recorded operator cases; supplementary full-model backward measurements assess integration into community FLA models and a GDN2 layer port. Code and reproduction materials: https://github.com/flashserve/ReWY.

0.28

Conditional Calibration Contracts under Coupled Load: A Branch-Aware Audit of Switched Inference

cs.LG cs.CLcs.DC ·2026-11-01 ·Juan S. Santillana
Abstract · 12 pages including references and appendices, ACM sigconf. Historical 40-cell campaign (10 training seeds x 4 coupling levels), with separat…

Bursty serving changes both queue states and the population selected for each model. We study conditional calibration contracts as a specification to be validated, rather than a certified property of an existing router. CA-MHD combines load-coupled distillation, closed-loop replay and a cost-informed Schmitt router in a discrete-event simulator with a hardware service table. Across 40 cells (ten training seeds and four content-load coupling levels), we distinguish the original teacher, frozen wide heads on an adapted trunk, and the narrow student. A common teacher-trajectory audit has median absolute student-teacher deficit 0.06 percentage points, but does not certify the student-selected deployment support. Historical results use a kernel correctness gap, not the published SmoothECE; we separately prove selected-branch composition for a residual functional at a common fixed scale. The deployed wide branch has lower ordinal accuracy than the original teacher in all 40 cells. Same-checkpoint student-only serving has lower historical ordinal gap in 27/40 cells; routing also exposes violation tails up to 17.5 percent. Student p99 latency remains above the 1.5 ms target. Retrospective temporal recalibration improves the ordinal gap in 39/40 cells, but causal delayed-label availability is unverified and no incremental per-mode benefit is detected after multiplicity adjustment. An in-sample CUSUM detects 30/40 checkpoint changes at five historical false alarms, with no prospective false-alarm guarantee. These findings support a branch-aware audit and a scoped negative result, not demonstrated routing superiority. Independent budgets, selected-support validation and net utility under a satisfied SLA remain open.

0.28

Hinted Scaling Laws

cs.LG cs.AIcs.CL ·2026-11-01 ·Suze van Adrichem, Luke Bailey, Rohith Kuditipudi, Christine Ye, Neil Band +4 more
benchmark
Abstract

Scaling laws predict the outcome of large, expensive training runs by extrapolating trends from smaller, cheaper runs. On downstream accuracy evaluations, these predictions fail when the weaker models produced by the cheaper runs score near zero accuracy, because these scores give no trend to extrapolate to larger scales. We introduce hinted scaling laws, which use privileged information to turn a benchmark into a continuum of easier versions so that even the small models score above zero. We find that accuracy rises smoothly and predictably with the hint level, enabling new types of scaling laws that operate below the noise threshold. On the Delphi pretraining ladder, a hinted scaling law fit on models trained with at most $10^{20}$ FLOPs forecasts the GSM8K accuracy of a $10^{23}$ FLOP model (25.9%) to within 3 percentage points; a standard scaling law under-forecasts it by 13 percentage points when fit on the same models and by 12 even with models up to $10^{22}$ FLOPs in the fit. Across both this pretraining ladder and a collection of post-trained model families, hinted scaling laws reduce the median forecasting error on GSM8K, AIME and HLE from 21 percentage points to 4.3. Hinting also reduces the number of test examples needed to tell whether one model is better than another by widening the accuracy gaps between models while approximately preserving their order: a single hinted rollout per problem ranks nearly indistinguishable models as accurately as five unhinted rollouts. Our results suggest that hinting is a general primitive for improving the accuracy and cost of downstream scaling laws and other prediction tasks in model development.

0.28

Benign population landscapes do not transfer to empirical risk minimization in nonsmooth optimization

cs.LG math.OC ·2026-11-01 ·Penghao Wang, Mateo Díaz, Liwei Jiang
theory
Abstract · 71 pages, 2 figures, 1 table

Benign landscapes provide a geometric explanation for the success of iterative methods on nonconvex problems arising in data science. The landscape is benign if every local minimizer is global and every other critical point is a strict saddle. For methods that avoid strict saddles, convergence to a critical point therefore implies convergence to a global minimizer. For smooth stochastic optimization, a common strategy for establishing such landscapes proceeds in two steps: $(i)$ establish benign geometry for the population objective---which is often easier to analyze---and $(ii)$ transfer this geometry to the empirical objective using concentration of gradients and Hessians. Recent work has extended saddle-point avoidance guarantees to weakly convex nonsmooth optimization through the notion of active strict saddles. We show, however, that benign population geometry need not transfer to empirical objectives in this setting. Specifically, for robust phase retrieval with Gaussian measurements---a well-studied and highly structured problem---the population objective is benign. Nevertheless, the empirical objective has sharp spurious local minima with probability bounded away from zero as the sample size tends to infinity. This asymptotic probability lower bound is independent of the dimension.

0.28

Abstention Spends Its Budget on Local Label Variance More Than on Novelty: Auditing Conformal Calibration and Selective Prediction under Molecular Scaffold Shift

cs.LG stat.ML ·2026-11-01 ·Arnav Srivastava, Krishna Kassetty
Abstract · 12 pages, 4 figures, 6 tables

Selective prediction with split conformal calibration promises a post-hoc fix for molecular property prediction: abstain on the least reliable compounds, ideally out-of-distribution chemistry, and the retained set becomes trustworthy without retraining. We audit this recipe for a gradient-boosted fingerprint model on lipophilicity (regression) and BBBP (classification) under a Bemis-Murcko scaffold split, holding the fitted model and evaluation molecules fixed while varying only the calibration fold (18 replicates). (i) On lipophilicity, coverage depends on calibration provenance: calibrating on held-out training-scaffold molecules falls below nominal (PICP 0.841 at a 0.90 target), while novel-scaffold calibration recovers it (0.904); on BBBP, coverage stays above nominal in every regime. (ii) For split conformalized quantile regression (CQR) the conformal correction is a constant, so conformalization does not change what is abstained on. (iii) The abstention score (CQR width; effectively ensemble variance on BBBP) tracks local label variance more than chemical novelty: controlling for local label variance largely explains away the novelty association (partial Spearman 0.067 and 0.047). (iv) At a fixed assay budget, gating on interval width or ensemble variance never improves hit precision over ranking by point prediction, at the pre-specified operating point or across 20 exploratory settings; a plain novelty gate shows small, uncorrected gains at the smallest budgets. (v) In an exploratory analysis, exact conformal selection exceeds its false discovery rate target under training-scaffold calibration (0.303 at q=0.2) and is consistent with it under novel-scaffold calibration, contributing a data-driven batch size rather than a better ranking. Conformalization earns its keep through guarantees whose validity hinges on calibration provenance, not by choosing what to abstain on.

0.28

Sensitivity Inflation from Cross-Record Precision Routing in Private Quantized Fine-Tuning

cs.LG ·2026-11-01 ·Annan Zhang, Shengjie Song, Wenshuai Yao, Lihao Zhang, Tianmiao
Abstract · 13 pages, 5 figures

Private fine-tuning of large language models increasingly pairs DP-SGD with low-rank adapters on frozen, quantized backbones, and every such stack inherits one unstated premise: frozen layers keep each record's clipped gradient a function of that record alone. We show that LLM.int8 breaks this premise. It routes outlier activation columns to a higher-precision path using a union over the whole batch, so one record decides how every other record's activations are computed, even though the weights are frozen and the quantization scales are per row. We bound the resulting sensitivity inflation, state a record-isolation contract that restores the standard bound, and define record-owned precision dispatch that meets it with the same GEMMs as batched LLM.int8, so isolation needs no extra arithmetic. We then audit the native CUDA stack (bitsandbytes, PEFT, Opacus) against four matched controls. A constructed LoRA-Opacus witness reaches $15.0\times$ the add/remove bound. On pretrained OPT-125M with natural text, other records alter 20% of record-layer precision splits at batch size 16. An exploratory re-clipping analysis finds 10 of 64 candidate neighbors above the bound once every gradient is clipped; the top pair, revalidated natively, reaches $1.065C$ while every control stays within $C$, and forcing one layer's outlier mask removes the excess. A two-million-draw audit of the released adapters yields $\varepsilon$ lower bounds ($3.25$ at $δ=10^{-5}$) below the advertised guarantee ($4.38$). Resolving an excess this small at the release level would take about 200 times more draws, whereas checking the sensitivity contract directly exposes it in a few native runs, which makes contract checking the practical audit for private quantized training.

0.28

RubricWorld: World Model-Guided Rubric Optimization for LLM Graders

cs.AI ·2026-11-01 ·Hongji Pu, Hang Li, Yucheng Chu, Jiliang Tang
Abstract · 25 pages including references and appendices; 6 figures, 8 tables

Rubric optimization improves LLM grading by revising scoring instructions while keeping the grader's parameters fixed. Existing optimizers often judge edits by aggregate validation performance, limiting the supervision they retain about which errors an edit corrects or introduces. Testing each candidate also requires rescoring validation answers, which limits how many edits can be explored within a fixed budget. We propose RubricWorld, a framework that learns from previous edit evaluations to guide further search. Its world model predicts an edit's correction and degradation rates at score boundaries, conditioned on the current rubric state. The training signal comes from paired grading results: scores for the same answers before and after an edit. Predictions select one candidate for direct evaluation, and held-out confidence bounds determine whether it corrects enough target errors while keeping new errors within prespecified limits. Measured effects update the model even when the edit is rejected. Across EIR (National), ASAP2.0, and ASAP-SAS with three frozen LLM graders, RubricWorld achieves the highest score in 20 of 27 metric comparisons. On ASAP2.0, it raises gpt-5-mini accuracy from 18.20% with the original rubric to 46.70%, exceeding the strongest baseline, APO, by 22.60 percentage points. code will be released upon acceptance.

0.27

When Disagreement Labels Do Not Identify F1 Rankings

cs.LG ·2026-11-01 ·Roberto Fernández-Barrios, Iker Pastor-López, Asier González-Santocildes, Pablo García Bringas
Abstract · 6 pages, 1 figure, 1 table. Submitted to ESANN 2027, submission ES2027-34. Code and reproducibility artifact: https://doi.org/10.5281/zenodo…

Agreement labels cancel from an accuracy comparison but can change an F1 ranking. We specialize partial identification to an explicit audit after labeling disagreements: two agreement counts give exact bounds, certifying a ranking or a choice within a specified F1 tolerance under every completion. Across 3,200 frozen model pairs, shared completions tighten bounds and a 0.01 tolerance reduces labeling costs. A zero-loss (weak-winner) control attributes most savings to positive tolerance, not ties. Guarantees concern the evaluated batch, not future data.

0.27

Hippocampal consolidation without replay in fully fine-tuned networks

cs.AI ·2026-11-01 ·Keuntae Kim, Yong Suk Choi
code/data
Abstract · 37 pages, 5 figures, 1 table; Extended Data and Supplementary Information included. Code: https://github.com/kkt94/deltahippo

Networks that keep learning overwrite what they know. Prevailing remedies replay stored data, distil from old models or freeze the backbone. The brain suggests a different division of labour: the hippocampus indexes where experience lies in cortex while the cortex keeps learning. We show that this division lets a fully trainable network keep learning without cross-task replay, a teacher or inference-time memory. After each task, a training-only hippocampal memory indexes the input geometry of every group of synapses sharing an input, residual energy included, and limits each synapse's movement per direction before and after the optimizer. This class-free geometry carries most of the protection; with conflict detection on top, the method cuts absorption of established classes by near-duplicates several-fold against every compared method. To our knowledge, this is the first demonstration of one principle keeping whole networks learning across four architecture families and models up to 8 billion parameters. Methods that freeze the backbone hold up only as far as its representation already separates the classes; here the whole network keeps learning to tell near-duplicates apart, leading the frozen-backbone baseline by 17.5 points on TACRED. Without the memory, much old knowledge stays decodable but beyond the network's reach.

0.27

RATSIFT: Resolution-Adaptive Transformer for Solar Irradiance Forecasting and Temporal Downscaling

cs.LG stat.APstat.ML ·2026-11-01 ·Jesto Peter, Timothy M. Hansen, Hossein Moradi Rekabdarkolaee
Abstract

Accurate forecasting and temporal downscaling of solar irradiance is critical for reliable renewable energy integration and grid management. Traditional methods often struggle to capture rapid irradiance variability. In this work, we develop a Transformer-based architecture for simultaneous forecasting and temporal downscaling. The model takes input data at arbitrary resolutions and generates outputs at any resolutions represented within the training data, leveraging both global and local contextual information. To improve training efficiency, a genetic algorithm-based optimization framework is implemented to identify suitable model hyperparameters, while reducing the computational cost of exhaustive grid search. The effectiveness of the proposed model is shown by analyzing real data in Dallas, Texas. The proposed model achieved root mean square error (RMSE) values of 28.785, 29.544, 30.157, 32.132~W/m\(^2\) for 15-, 30-, 60-minute, and irregularly spaced global input and producing forecast at 5-minutes resolution, respectively. Furthermore, this model produced reliable uncertainty estimates, exhibiting wider prediction intervals for high-resolution inputs during volatile solar conditions, whereas coarser-resolution inputs result in narrower intervals and significantly lower training time and computational cost. Our approach provides a scalable and efficient framework for multi-resolution solar irradiance forecasting and temporal downscaling, suitable for practical energy forecasting and grid management applications.

0.27

The (Butter)fly Dream: Linear Separability of World-Model Rollouts in a Connectome-Constrained Visual Model

cs.LG q-bio.NC ·2026-11-01 ·Junichiro Niimi
unusual
Abstract

Zhuangzi dreamt he was a butterfly and, on waking, could not tell which was the dream. We ask a version of his question of a model of the fly brain: placed inside a world generated by a world model, can a fixed brain tell that the world is a dream? We treat it as a question about what a connectome-constrained model trained on one task makes available beyond that task: do its responses make the defects of a dream linearly decodable? The observer is a published model of the fly visual system trained only for optic-flow estimation; reality is the MPI-Sintel movie on the model's 721-column retina; dreams are continuations of five real frames by three small world models (a persistence random walk, an MDN-RNN and an RSSM) at three temperatures. The judge is a classifier two-sample test on temporal statistics of cell-type responses, with sequence-level cross-validation. In the trained observer's responses all nine dreams are separable from reality (AUC 0.85-0.99, mean over 50 trained models), in those of the same network before training much less so (0.58-0.75 for the learned world models; paired difference 0.20-0.37), equally on sequences the observer never saw, and in an order unrelated to pixel error. Because these AUCs are near ceiling, we add graded probes made of each window's own real frames with only their order broken. Swapping one adjacent pair of the 14 generated frames gives 0.73 on the trained responses against 0.54 before training and 0.56 for pixel statistics; the trained curve reaches 0.75 at 1.1 swapped pairs of seven, the pixel statistics at 4.7, and the untrained network not at all. A detection horizon, the test on the first k generated frames, orders the world models where the full-length AUC saturates.

0.27

Causal Optimal-Transport Attention: Positional Priors and Length Extrapolation

cs.LG ·2026-11-01 ·Ming Lian
code/data
Abstract · 42 pages, 5 figures; code: https://github.com/homw314159-spec/causal-ot-attention

Causal masking obstructs the column balancing used by optimal-transport (OT) attention. We study an unbalanced OT operator that uses only current and earlier rows and characterize its induced positional prior. For bounded scores and any fixed balancing strength below one, its fixed point assigns vanishing mass to every fixed window as context grows. With constant scores, the prior has an exact scale-invariant form. We classify positional encodings by whether they prevent this dilution. We also derive finite-length window-mass bounds for log-distance priors with exponents near one. Preregistered experiments use 25M- and 100M-parameter language models. The OT operator improves loss extrapolation over softmax, while exploratory comparisons favor an explicit distance bias at a fraction of the cost. Changing only the bias beyond the training length moves long-context loss in the predicted direction on all 30 tested checkpoints. During training, learned exponents approach one even when constrained above it. A margin chosen from prior calculations controls distant-mass growth in the tested setting.

0.27

Stand-In Effects and Pooling Amplification: Lessons from Biological World Models for Testing Action Effects

cs.LG q-bio.QMstat.ME ·2026-11-01 ·Bowen Xu, Boyu Chen, Jiaying Li
Abstract · v3: adds a second registered test on two further published world-model families (IRIS, TD-MPC2; OSF j75g2), exploratory studies of score rul…

Action-conditioned world models are increasingly trained on data from randomized experiments, and whether their predictions respond to actions is usually checked by replacing or removing the actions and measuring the loss in accuracy. Taking biological experiments as the guiding case, we show that such diagnostics report action effects where there are none: under autocorrelated designs, future actions stand in for history the model does not see. When the design law is known, a conditional randomization test written as an e-value is valid in finite samples whatever the model, provided the resampler conditions on the full state the design used. A second failure arises whenever many sites share one action sequence, as in world models: the pooled score turns small model errors along the action contrast into overwhelming evidence against a true effect. In a pixel world, its median log e-value fell from -17.3 to -6832 as frames grew from 64 to 4096 pixels, while the oracle's rose. Tempering the score changes the bet, not the validity. Our own preregistered analysis of optogenetic ERK experiments failed in exactly this way; a tempered statistic, registered before the data were obtained, then rejected the null on a never-opened field (E = 1.1e10). Every single tempering factor we tried, estimated from residuals or tuned on validation data, collapsed in some setting; mixtures over a fixed grid of factors never did. In registered tests on three world-model families trained by others, the pooled test rejected in 0 of 96 units of a diffusion model in which actions mattered, and reached the registered power bar in 10 of 26 Atari games under IRIS's own token likelihood and 1 of 6 control tasks under TD-MPC2; mixtures fixed in advance reached it in all 96 units, 24 of 26 games and 6 of 6 tasks. Under exact nulls the statistics made 4 rejections in 22,848 tests.

0.27

TIGER: Learning from Researcher Taste to Generate Tailored Research Ideas

cs.AI ·2026-11-01 ·Shane Gu, Yufeng Lin, Dixi Yao, Xuchen Gong, Chenhao Tan
unusual
Abstract

In pairwise comparisons where each judge chooses between an idea generated by its corresponding TIGER-8B policy and one generated by GPT-4.1, the three LLM researchers select the TIGER-8B idea in 56.6%-68.4% of 76 comparisons per researcher. PhD1 and PhD2 select the TIGER-8B idea in 19 of 30 pairs (63.3%) and 17 of 30 pairs (56.7%), respectively. All six directed cross-policy comparisons likewise favor the judge's matched TIGER-8B policy over an off-target policy, supporting the use of individual preference feedback for tailored research ideation.

0.27

An auditable agentic pipeline for variant-to-gene prioritisation, validated against orthogonal functional evidence

cs.AI q-bio.GN ·2026-11-01 ·Shakuntala Baichoo, Sohail, Sarkar, Bo Wang
benchmark
Abstract · 21 pages of main manuscript including References, 4 figures, 11 supplementary tables, 2 main tables

Assigning GWAS variants to causal genes requires evidence from multiple resources. We present V2G, an agentic pipeline that records evidence retrieval and returns ranked genes alongside the retrieved evidence. We evaluate whether language-model reranking improves gene prioritisation and whether its gains depend on the retrieved evidence or recall of known associations. On 135 held-out loci where candidate retrieval succeeded and Open Targets L2G predicts, reranking by a language model improves Precision@1 over the evidence ordering it receives, 0.793 against 0.711 (+0.081, 95% CI +0.038 to +0.129), and over nearest-gene assignment (+0.178); no difference from Open Targets L2G or from training-membership baselines was detected. Given only the trait and the candidate gene symbols, in shuffled order, the same model scores 0.864 on these loci (three seeds, 0.859 - 0.867) and names the reference gene at 0.594 of the loci where the pipeline never retrieved it; with each gene's symbol moved to another gene's evidence row it follows the top-scored row and scores 0.284. The gain over the evidence ordering is recall of known genes, and the evidence as presented anchors the model to the ordering rather than informing it. Across the 296 held-out loci (five seeds each) the model keeps the given top candidate at 86% of decisions; at naturally retrieved loci its departures select a reference gene in 85% of departure decisions (31 of 37 loci), and none displaced one. Its calls carry trait-matched mouse mutant or Mendelian phenotype annotations more often than other candidates at the same locus. Train-test overlap is reported throughout. The study includes a frozen benchmark, cached predictions and reproduction code.

0.26

A Self-Generated Floor on Adam's $\varepsilon$: Evidence from Power-Law Matrix Sensing

cs.LG ·2026-11-01 ·Lngel Kyo
Abstract · 23 pages, 3 figures, 5 tables

In nonconvex Burer--Monteiro low-rank matrix sensing of $U^\ast \in \mathbb{R}^{V \times D}$ under power-law covariance $Σ= \mathrm{diag}(j^{-s})_{j=1}^V$, the precision-wall approximation predicts that Adam stalls when the smallest non-gauge Hessian eigenvalue drops below its denominator floor, giving $s^\ast(V,D) = [\log(1/\varepsilon_{\mathrm{eff}}) - \log V]/\log(VD) + δ_{\mathrm{Adam}}$. If $\varepsilon_{\mathrm{eff}}$ were set by $\varepsilon_0$, lowering $\varepsilon_0$ from $10^{-10}$ to $10^{-12}$ would raise $s^\ast$ by $0.95\text{--}1.33$. Instead, across six values of $\varepsilon_0$ two decades apart, $10^{-2}$ to $10^{-12}$, the boundary follows the $\varepsilon_0$ slope down to $10^{-6}$, bends near $10^{-8}$, and is flat below $10^{-8}$, while warm-started L-BFGS and Gauss--Newton reach machine precision on stalled instances. We trace this saturation to a self-generated dynamical floor $\varepsilon_{\mathrm{dyn}} \propto α$: high-curvature modes oscillate persistently near the adaptive edge of stability with amplitude $\propto α(1-β_1)/(1+β_1)$ and leak gradient fluctuations of scale $κα$ into rare modes via measurement coupling, pinning $\sqrt{\hat{v}_V} \approx κα\gg \varepsilon_0$. Fitting a single threshold-crossing wall parameter $\varepsilon_{\mathrm{dyn}} = 4.76\times 10^{-9}$ (with the detection lag frozen at its companion calibration) predicts the saturated failure boundary for $D \ge 4$ across five $(V,D)$ scales to RMS $0.098$ (breaking down at $D = 2$), while shrinking $\varepsilon_0$ below $10^{-8}$ is ineffective and lowering the step size $α$ reduces the floor directly.

0.26

Fuzzy-Enhanced Machine Learning Approach for Type 2 Diabetes Diagnosis

cs.LG ·2026-11-01 ·Azam Saffar, Mohsen Saffar
published
Abstract

Accurate diabetes prediction is critical for early clinical intervention, yet existing machine-learning pipelines often fail due to class imbalance, noisy physiological measurements, and redundant or weakly informative features. To address these unresolved limitations, this study establishes a unified fuzzy-evolutionary-ensemble framework designed to uncover a more discriminative feature space for type 2 diabetes prediction. The workflow integrates selective fuzzification of nonlinear clinical variables, SMOTE-ENN-based balancing and noise filtering, and Bat-Algorithm-driven evolutionary feature selection prior to training a Random Forest classifier. Applied to the Pima Indians Diabetes Dataset, the framework compresses the expanded fuzzy-numerical space into an optimized 10-feature subset and yields substantial performance gains: 96.11% accuracy, 95.14% precision, 98.00% recall, and a 98.23% F1-score. Relative to a baseline Random Forest using all 16 features (74.45% accuracy; 59.86% F1), the system achieves more than a 1.6-fold improvement in F1-score while preserving over 99% stability in majority-class detection and surpassing earlier published 16-feature models impacted by imbalance and weak feature relevance. Collectively, these results demonstrate that the integrated pipeline produces an enhanced fuzzy-numerical representation space that improves diabetes risk stratification and provides a scalable, uncertainty-aware foundation for diagnostic modeling across broader clinical datasets.

0.26

Beliefs in feedback: Factored representations of interacting agents in transformers

cs.LG cs.AIcs.MAnlin.AO ·2026-11-01 ·David A. Ehrlich, Manuel Baltieri, Fernando E. Rosas, Alexander B. Boyd
Abstract

Complex systems, from ecosystems to brains, are usually modeled as modular components coupled through feedback rather than as indivisible wholes. Can a transformer trained on the joint behavior of a group of interacting agents be understood in the same way? In this work, we develop a computational-mechanics framework to study agents interacting in feedback loops. We formally prove that any joint stochastic process can be presented as resulting from the behavior of agents interacting in feedback, and the resulting process is generated by a hidden Markov model whose Bayesian beliefs factorize into separate components per agent. Furthermore, our experiments show that transformers trained on such feedback sequences make use of this separability, disentangling the beliefs about the underlying factors into separate subspaces of their residual stream. This representation is found to scale linearly (instead of exponentially) with the number of agents. Moreover, interventions show that these representations are causally efficacious, allowing for targeted steering of individual agents' beliefs by selectively inserting counterfactuals. However, coarse-graining the behavior is found to disrupt the factorization, and can force a higher-dimensional representation. Together, the results establish both the dimensional advantage of a factorized representation and the ability of transformers to exploit it, laying the groundwork for investigating traces of compositional predictive architecture in pretrained models.

0.26

Mesh2Shell Neural Operator for Inverse Shape Optimization

cs.LG ·2026-11-01 ·Xinyi Li, Xi Deng, Yubo Cao, Hudson Liu, Anima Anandkumar
Abstract

Physics-driven design seeks high-performing geometries, which need not conform to a learned shape category. Parametric and latent-space approaches facilitate surrogate-based optimization, but restrict deformations to prescribed shape families or learned decoders. We introduce the Mesh-to-Shell Neural Operator (M2S-NO) as the mesh-to-shell component of a category-agnostic inverse shape optimization framework. We separate the explicit mesh used for design from the volumetric representation used for physics prediction. M2S-NO maps oriented surface samples to a surface-centered shell representation, such as TruncatedSDF or HeatSDF. A separate problem-specific PDE neural operator (PDE-NO) maps this representation to physical fields. Objective gradients update mesh vertices through both operators, without a latent shape decoder. Preconditioning and remeshing support free-form deformation, while fixed interfaces are controlled directly on the mesh. Optional shell-space proposals enable topology changes. We evaluate geometric matching, jet-engine brackets, and neutron moderators, with independent FEM and OpenMC evaluation. Relative to the best training designs, M2S-NO reduces maximum bracket displacement by 14.3% and increases neutron-sensor current by 13.8%. Optimization also succeeds from an out-of-distribution bracket initialization, demonstrating the framework's geometric flexibility.

0.26

A Hybrid Anomaly Detection Method for Electric Vehicle Identification from Smart Meter Recordings

cs.LG cs.AI ·2026-11-01 ·Ali Babalhavaeji, Hui Song, Mohammadreza Radmanesh, Mahdi Jalili
Abstract

In recent years, electric vehicles (EVs) have attracted much attention due to their potential to decarbonize the transport system. EV charging imposes challenges on power grids, such as load imbalance and voltage instability. Therefore, electricity distributors need to identify which users have an EV. Smart meter recordings can be effectively used for this purpose. The major issue associated with such data is the imbalance problem, where the majority of the data belongs to users without EVs. The existing methods for EV identification from smart meter recordings have framed it as a classification task. In this paper, we frame the problem as an unsupervised problem and propose a hybrid anomaly detection method that combines the power of prediction-based and reconstruction-based anomaly detection methods. The proposed model considers temporal and stochastic representation information of smart meter time series via prediction-based and reconstruction-based components, respectively. The performance of the proposed model is compared with state-of-the-art supervised and unsupervised methods. Its superiority is verified on real-world data.

0.26

Uncertainty-Aware 3D Position Reconstruction in CZT Drift-Strip Detectors

cs.LG ·2026-11-01 ·Alejandro Valverde Mahou, Michał Kossakowski, Selina R. H. Owe, Rikke S. Klausen, Irfan Kuvvetli +2 more
Abstract

Accurate three-dimensional interaction-position reconstruction is important for CZT drift-strip detectors, but conventional methods can suffer from degraded performance in challenging detector regions, while learned models provide limited information about prediction reliability. We investigate two uncertainty-aware approaches: deep ensembles of point-estimation networks and simulation-based inference (SBI), which estimates a posterior distribution over interaction position. Both approaches are trained on physics-based simulations and evaluated on both synthetic and experimental data. On synthetic data, uncertainty from both methods correlates with positioning error, with SBI showing a substantially stronger relationship. On experimental measurements, the deep ensemble achieves improved $z$-position resolution, while SBI provides more informative uncertainty estimates that highlight regions of reduced positional information and discrepancies between simulation and the physical detector. These results demonstrate the complementary strengths of ensemble-based reconstruction and posterior inference for uncertainty-aware detector position reconstruction.

0.26

ANCHOR: KL-Anchored Adaptation for Stable Offline-to-Online RL

cs.LG cs.AI ·2026-11-01 ·Geonwoo Cho, Jaegyun Im, Doyoon Kim, Lexin Li
code/data
Abstract · Accepted at the CoRL 2026 Workshop on Continually Self-Improving Robots. 37 pages

Offline-to-online reinforcement learning (RL) follows the pretrain-finetune paradigm: an agent is first pretrained on historical data and then adapted through online interaction. We study offline-to-online adaptation without offline replay, where the agent must improve using only the pretrained model and newly collected experience. A key challenge is to avoid catastrophic failure, a sharp early performance collapse during online fine-tuning, while still achieving strong asymptotic performance. Through controlled experiments, we show that removing offline replay can induce catastrophic failure, while retaining strong conservative regularization from offline RL can suppress final online performance. We also find that common stabilization strategies do not reliably prevent early collapse, and that the critic-side mechanisms we test are insufficient on their own. Motivated by these findings, we propose ANCHOR, a simple adaptation method that anchors the actor to the offline-pretrained policy early in online fine-tuning, anneals this constraint over time, and retains only a small residual conservative critic regularizer. We provide a theoretical interpretation showing that KL anchoring helps preserve critic reliability across nearby occupancy distributions, with a two-phase regret bound explaining when anchoring should be maintained or relaxed. Across FrankaKitchen, AntMaze, and Adroit tasks, ANCHOR substantially reduces catastrophic failure while maintaining strong asymptotic performance. Ablations show that KL annealing and the residual conservative weight are important for balancing early stability and online adaptation, while additional analysis shows that ANCHOR is most beneficial when the pretrained policy contains useful behavior to preserve. Code is available at https://github.com/Cho-Geonwoo/ANCHOR.

0.26

LARK: Lean-Assisted Reasoning for LLM-Generated GPU Kernels

cs.AI ·2026-11-01 ·Zixi Zhang, Jiayi Nie, Haoran Wu, Yujie Yan, Yann Herklotz +5 more
code/data
Abstract

Efficient GPU kernels are typically low-level and hardware-specific, and LLM agents can now generate them faster than experts can verify. Existing approaches rely on testing, which provides evidence but not a guarantee of correctness. We introduce LARK, an agentic framework that automatically and formally verifies GPU kernels against their PyTorch reference implementations. LARK proves equivalence in functional and memory behavior under exact arithmetic and establishes a new trust boundary: human effort is only needed to verify operator-level Lean semantics ahead of kernel generation time, instead of checking every single line of generated code. LARK formalizes the GPU machine model and the semantics of each operation in Lean 4 once per target language, and the Lean 4 semantics of any incoming kernel is derived from these pre-built definitions. LLM agents then automate the proof in two steps. First, an agent decomposes the whole kernel semantics into smaller segments and Lean 4 verifies this segmentation, which addresses the intractability of proving equivalence for large kernels directly. Second, agents write Lean proofs that each kernel segment matches its high-level PyTorch counterpart. Together, the two steps give whole-program equivalence. We demonstrate that LARK can be used 1) to ensure correctness in existing code-generation agents and 2) to automatically verify human-written kernel libraries. On the KernelCraft GPU tasks, combined with a simple ReAct agent, LARK enables GLM 5.3 to fully verify 10 of 12 Level-1 kernels and Opus 5.5 all 5 Level-2 kernels. Moreover, over 10 production kernels randomly sampled from popular open-source LLM serving frameworks, LARK verifies 9 and identifies a new data race bug in the tenth. We believe LARK represents a significant step toward verifiable code generation in the agentic era.

0.25

Autoencoder Based Deep Clustering of Point Clouds

cs.LG cs.AI ·2026-11-01 ·Nader Alfares, Guangmingmei Yang, David J. Miller, George Kesidis
Abstract

Motivated by a problem in Generative Specification-Producing AIs (SPAIs), we focus on a problem of deep clustering of three-dimensional shapes specified by point clouds. After reviewing clustering and deep clustering, we study the fully unsupervised pipeline in which a FoldingNet autoencoder is first trained for reconstruction and then K-means++ is then applied to the resulting codewords. We report clustering accuracy (ACC), normalized mutual information (NMI) and adjusted Rand index (ARI) on ModelNet10 and ModelNet40 datasets, as a function of the number of clusters K. We also provide t-SNE visualizations of the codewords, and discuss the choice of K.

0.25

Privacy-Preserving ECG-HRV Emotion Classification for Mood Swing Detection using Federated Learning on Constrained Edge Devices

cs.LG cs.AI ·2026-11-01 ·Chidike Vincent, Muhammad Aslam Jarwar, Najam Ul Hasan, Laurence Hirsch, Jonathan Zasada-James +1 more
Abstract

This paper presents a privacy-preserving mood monitoring framework that classifies emotional states from ECG-derived heart rate variability (HRV) features. The framework is evaluated on three public datasets comprising 74 participants and further validated using a wearable device in a pilot study. Leave-One-Participant-Out cross-validation (LOPO-CV) is used to assess generalization to unseen individuals. A series of experiments investigates the impact of temporal windowing, HRV feature selection, and class balancing on both four-class valence-arousal and binary valence classification. Results show that class balancing substantially improves minority-class recall, while Logistic Regression consistently outperforms five alternative classifiers and achieves competitive performance against HRV-based baselines. Pilot validation using a Polar H10 wearable achieves 75% binary valence accuracy on an unseen individual without calibration, demonstrating the feasibility of real-world deployment. To preserve privacy, the framework is also implemented using federated learning on a six-node Raspberry Pi cluster. Federated Logistic Regression achieves performance comparable to centralized training without sharing raw physiological data. These results demonstrate the potential of interpretable and privacy-preserving mood monitoring for continuous mental health assessment in clinical and community settings.

0.25

Sweep Less, Steer More: Using Activation Steering Geometry for Reduced Hyperparameter Search

cs.LG cs.AIcs.CL ·2026-11-01 ·Nishkal Hundia, Swastik Agrawal, Navita Goyal, Sarah Wiegreffe
Abstract

Representation steering methods like difference-in-means (DIM) offer a lightweight approach to controlling LLM behavior, but their effectiveness varies widely across concepts, datasets, and layers, making principled deployment costly without exhaustive hyperparameter sweeps. We build on the discriminability index $d'$, a compute-cheap geometric measure of how well contrastive training activations separate along the steering direction. Across five behaviors on 3 models, we demonstrate uses beyond its established role as a dataset-level steerability predictor. First, in the multiple-choice question answering (MCQA) setting, we show that $d'$ reliably identifies performant layers at which to steer, eliminating the need for a layer-wise sweep. Second, we extend this analysis to open-ended multi-token generation, and find that $d'$ remains successful for layer selection across different settings. Lastly, we test whether the position of a generated token's representation on the DIM line predicts steering success on that individual example. In the MCQA setting, this works for high-$d'$ datasets but is inconsistent across behaviors, models, and training-token choices in open-ended generation. This highlights that methods developed and conclusions drawn in single-token MCQA settings do not necessarily transfer to open-ended generation.

0.25

Surrogate Losses for Fairness-Aware Classification: Fisher Consistency and Calibration

cs.LG stat.ML ·2026-11-01 ·Changcheng Song, Yi Yang, Xiangyu Chang, Xiaoling Lu, Yufeng Liu
hot direction
Abstract

Fairness-aware classification seeks to balance predictive accuracy and fairness across demographic groups. Although surrogate losses facilitate the optimization of fairness-aware objectives, it remains unclear when minimizing these losses yields optimal decisions under the original fairness-aware risk. This paper investigates the statistical validity of surrogate-based fairness-aware classification under demographic parity and equal opportunity, considering both attribute-blind and attribute-aware settings. We introduce fairness-aware Fisher consistency and characterize it through conditions on the derivatives of the surrogate losses. The characterization shows that consistency depends not only on the individual loss functions but also on their interaction within the fairness-regularized objective. We further establish a calibration inequality that relates the excess fairness-aware risk to the excess surrogate risk, providing a quantitative connection between surrogate optimization and the original classification objective. Numerical experiments illustrate how deviations from the consistency conditions affect fairness-aware excess risk and how the regularization parameter influences the well-posedness of surrogate optimization. An application to consumer credit default prediction demonstrates that appropriately calibrated fairness regularization can substantially reduce demographic disparities while maintaining predictive performance.

0.25

Geometric Morphology Learning: Empowering Foundation Models from Structure to Spacetime Intelligence

cs.AI ·2026-11-01 ·Li Sun, Philip S. Yu
Abstract · 7 pages, 1 figure, TKDD 20th Anniversary

Geometry is the very form that connects space, time, and matter, serving as the fundamental fabric of the physical world, from continuous fields to discrete structures. Despite its importance, deep learning is inherently blind to geometry. Riemannian geometry stands as a well-established mathematical foundation for studying geometry, yet it remains largely incompatible with deep learning architectures. To bridge this gap, we establish a new learning framework--Geometric Morphology Learning--that learns geometric or topological structures, shapes or conformations for representation learning, reasoning and action modeling. Moving beyond early attempts limited to specific geometric priors, this theory centers on modeling the intrinsic geometry of objects and learning from such geometric or topological knowledge. Specifically, we construct a neural bundle, a unified scaffolding linking local and global structures, and formulate differentiable operators, including connection and curvature, thereby enabling the study of geometry with advanced geometric concepts such as holonomy and cohomology. Consequently, Geometric Morphology Learning opens new horizons in foundation models, reasoning and planning, graph modeling, physical simulation, PDE systems, and dynamics, marking a fundamental shift toward instilling systematic geometry in artificial intelligence.

0.25

Chain-of-Thought Oversight Requires Separate Evidence for Faithfulness and Monitorability

cs.AI cs.CR ·2026-11-01 ·Sichao Li, Sai Ma, Xiyang Hu, Chudi Zhong, Tongliang Liu
publishedposition
Abstract · The paper is accepted by Neurips 2026 Position Track

A common assumption in chain-of-thought (CoT) oversight is that a visible reasoning trace makes model behavior easier to monitor. We argue that this assumption can conflate distinct properties along a chain linking a model's internal computation, reasoning artifacts, outputs, and monitor judgments. We identify four properties along this chain: reasoning visibility (whether a trace is accessible), mechanistic faithfulness (whether a trace reflects internal computation), mechanistic localizability (whether a compact causal mechanism can be isolated), and behavioral monitorability (whether a monitor can detect the target behavior). Two complementary case studies across three open-weight reasoning models show why these distinctions matter. In the security vulnerability reasoning study, probes and activation ablations provide evidence of internal representation and causal influence, but a monitor with access to the reference answer does not outperform the majority-class baseline in a held-out test. In authority-hint factual reasoning study, a monitor correctly distinguishes changed from unchanged answers, but seven mechanistic analyses do not isolate a compact causal mechanism. These results show that evidence for one claim does not establish the others. We therefore propose an evidential-chain reporting standard. Each claim carries a licensing statement that specifies its target property, access tier, and artifact, and labels each required link as supported, unsupported, or assumed. These assessments determine whether the conclusion is licensed, conditionally licensed, or not licensed.

0.25

Can Large Language Model Agents Learn by Watching Tutorial Videos?

cs.LG cs.AIcs.CLcs.CV ·2026-11-01 ·Haoyu Zhao, Zafeirios Fountas, Jun Wang
benchmark
Abstract

Humans continuously acquire knowledge from carefully crafted materials such as books and tutorial videos, carrying that understanding into future tasks. Millions of such instructional videos are uploaded online every day, yet LLM agents remain unable to exploit this ever-expanding wealth of human teaching. Their capabilities are frozen after training, while parameter updates at scale remain prohibitively expensive. This raises the central question: can LLM agents learn from this vast instructional resource without parameter updates? We introduce Tutorial Agentic Learning (TAL), a nonparametric continual-learning paradigm that consolidates tutorial videos into a persistent, structured tutorialbook memory through a two-stage adversarial process that harvests hard, video-grounded questions and uses them to refine the memory. At inference time, TAL retrieves through a Table-of-Contents (TOC) tree, tying retrieval directly to the model's reasoning rather than to embedding similarity. To study this setting, we introduce VidLearn, a multimodal QA benchmark built from 72 chess tutorial videos spanning 12 openings. On VidLearn, TAL outperforms all evaluated baselines and improves monotonically as more videos are consumed, achieving up to 113% and 91% relative gains over the corresponding vanilla MLLMs on the training and test splits, respectively. In actual gameplay, TAL reaches a stronger opening-phase position than its vanilla counterpart in 69% of head-to-head games. Evaluation in a second domain further demonstrates the generality of TAL. Together, these results highlight instructional videos as a scalable resource for parameter-update-free agent continual learning.

0.25

Just-in-Time Skill Retrieval via Long-Horizon State Tracking

cs.AI ·2026-11-01 ·Xin Lin, Ziheng Wang, Zerui Chen, Jiayao Ma, Zhimin Wei +4 more
code/data
Abstract

Large language model (LLM) agents increasingly rely on external skill libraries to support complex task execution. However, determining which skills are needed remains challenging in long-horizon tasks. Existing methods typically select a fixed skill set before execution, implicitly assuming that a task's procedural requirements can be determined from its initial description. This assumption is restrictive for long-horizon interaction, where new requirements emerge from intermediate observations, execution outcomes, and newly revealed constraints. This motivates just-in-time skill retrieval, which dynamically selects skills as execution unfolds. Just-in-time skill retrieval is promising but challenging: the utility of a skill depends not only on the current context, but also on the evolving task state shaped by prior steps. To this end, we propose JitSkill, a state-grounded framework that addresses this challenge through continuous task-state tracking. By connecting emerging requirements with accumulated task progress, JitSkill provides procedural support throughout execution without treating each retrieval decision independently. Experiments across three agent benchmarks demonstrate its effectiveness in improving long-horizon task completion. Our code is available at https://github.com/XMUDeepLIT/JitSkill.git/.

0.24

Iterative Gumbel Planning for Continuous Control

cs.LG cs.AI ·2026-11-01 ·Shaohuai Liu, Weirui Ye, Yilun Du, Le Xie
top venue
Abstract · Accepted at NeurIPS 2026

Search-based planning remains a powerful policy improvement paradigm in reinforcement learning, driving milestone successes in discrete domains. However, extending this paradigm to continuous control exposes a structural mismatch: search evaluates a finite set of action candidates, whereas continuous RL typically relies on Gaussian policies. We show that winner-action distillation into a continuous Gaussian induces a \textit{projection error} and a stepwise variance-expanding gradient tendency when the selected action is sufficiently far from the current mean. Empirically, this tendency is associated with elevated terminal variance and a return gap on the evaluated tasks. To address this, we propose \textbf{Iterative Gumbel Planning (IGP)}. By adopting a per-dimension factorized categorical policy, IGP naturally aligns the policy support with the search candidates, eliminating the projection error and reducing the residual approximation to a bounded discretization error. To bypass the combinatorial explosion of high-dimensional discrete action spaces, IGP pairs Gumbel Top-$K$ candidate sampling with an MPPI-style multi-round iterative search to refine the policy at each state. Empirically, IGP achieves state-of-the-art aggregate performance and sample efficiency on DMControl and HumanoidBench. Furthermore, it produces more stable action sequences with lower temporal variance, demonstrating its practical potential for real-world robotic deployment.

0.24

Minimal Superior Penalties: Characterising and Calibrating Decision-Aware Proper Scoring Rules

cs.LG stat.ML ·2026-11-01 ·Rouhollah Ahmadian
theory
Abstract

Training a classifier means letting a validation metric decide which model is kept and when training stops. A natural requirement on that metric is superiority: every correctly classified forecast should score better than every incorrectly classified one, however close the underlying probabilities are. Neither the Brier score nor the logarithmic score satisfies this on its own. We show that superiority is governed by the separation between two endpoints -- the supremum of the score on correct predictions and the infimum on incorrect predictions -- and derive both in closed form for each score. Their difference is the infimal penalty, the greatest lower bound on the penalty needed to restore superiority. Prior work restores superiority with a penalty calibrated from the first endpoint alone; for the Brier base, our infimal penalty is smaller than that published penalty by a factor ranging from 4.00 at three classes to 2.11 at twenty. We further embed the published and infimal penalties in a one-parameter family and show that, under a single-valued tie convention, every member of the family is strictly proper. In a pre-specified two-dataset case study, mean selected-model macro-F1 rises with the penalty in the reported ten-class regimes, though not distinguishably from noise once seed/selection-mode nesting is accounted for; a fifth regime at forty-seven classes does not reproduce even this descriptive pattern. The intended no-superiority baseline is algebraically a rescaling of the family under test, so the experiment cannot attribute the observed pattern specifically to superiority. When superiority itself is the objective, any penalty strictly above the infimal value suffices; choosing a larger penalty is a separate calibration trade-off rather than a requirement for the guarantee.

0.24

A sensor-driven physics-informed neural network for real-time occupant-centric space heating control

cs.LG eess.SY ·2026-11-01 ·Yunfei Bai, Chenghao Li, Sunku Prasad Jenne, Hao Wang, Sizhe Zhang +1 more
Abstract

Conventional space heating systems typically rely on single or sparse fixed-point temperature measurements, overlooking non-uniform indoor temperature distributions, transient thermal dynamics, and occupant-specific comfort requirements. This study explores an occupant-centric control strategy that aims to maintain the desired temperature around the occupant location. Spatially and temporally varying temperature fields, together with occupant mobility, make real-time local temperature estimation challenging. To address this challenge, we propose a control-oriented Physics-Informed Neural Network (PINN) model for indoor temperature-field prediction. Conventional PINNs rely on explicit time inputs and assume consistent temporal patterns across samples, which limits adaptability under variable initial conditions and heating sequences. The proposed model infers temporal dynamics implicitly from spatial coordinates and limited sensor measurements, improving robustness to changes in initial conditions and external disturbances. The method was validated across nine horizontal sensing planes in three controlled room-scale configurations. Against a validated COMSOL Multiphysics simulation, the PINN predicted the evaluated temperature fields with a maximum relative error below 1.85% across validation cases. These predictions supported an occupant-centric space heating control scheme. Compared with COMSOL-based offline control, the PINN-based system achieved an average prediction error of around 0.67%, and its rapid inference supports real-time occupant-centric control implementation. This work represents a move away from reliance on fixed-point temperature measurements in space heating control to an occupant-centric control strategy informed by real-time environmental feedback.

0.24

FactorSpec: Serving Exact Relational Constraints as Speculative Drafts

cs.LG cs.AI ·2026-11-01 ·Jianchang Su, Dawei Xiang, Wei Zhang
Abstract · 18 pages, 14 figures, 5 tables

Speculative decoding speeds up LLM serving by drafting tokens that the target model verifies in one pass, and its gain grows with the drafts the target accepts. Structured outputs should be an ideal case, because much of a record follows from relations among its fields: an itinerary's days follow from its cities' stays and flights. Serving engines, however, draft such text by proposing it from the prompt or a learned head, and their grammar masks enforce only syntax, so the model must satisfy the relations, which it violates: on NATURAL PLAN's trip problems, Qwen3-8B writes a valid itinerary for 10%, and retries that name the violated relation raise this to 14%. We present FactorSpec, which serves exact relational constraints as speculative drafts. FactorSpec compiles a JSON Schema and its relations into a factor graph with an exact feasibility check, drafts every token the constraint determines as certain, verifies each request at a width chosen for it, and computes the masks in worker processes outside the scheduler while the GPU runs. In SGLang, its drafts yield up to 10.5 tokens per verify step, up to 1.9x those of the fastest n-gram drafter, and it decodes up to 1.43x faster than speculators tuned per load, at a grammar mask's cost; it also leads in vLLM. Because every record is also valid, it serves up to 13.9x the valid records per GPU-second of syntax-only baselines.

0.24

MIRA: Mutual Information guided calibration for Reliable Test-Time Adaptation

cs.LG ·2026-11-01 ·Hyeongyu Kim, Mingyeong Jang, Youngjun Song, GeonHui Han, Dosik Hwang
top venue
Abstract · Accepted at NeurIPS 2026

Test-time adaptation (TTA) updates a deployed model using only unlabeled target data, but standard entropy-based methods often rely on deterministic confidence and can reinforce overconfident errors under distribution shift. We propose MIRA, Mutual Information guided calibration for Reliable Test-Time Adaptation, a source-free online adaptation method that uses lightweight last-layer Gaussian perturbations to probe prediction reliability. For each target batch, MIRA reuses a single feature extraction pass and samples perturbed classifiers around the source classifier, forming an efficient local predictive distribution. Rather than fitting a Bayesian posterior, MIRA treats the perturbation scale as a local probing radius and selects a batch-specific scale using mutual information (MI) and prediction agreement. The resulting perturbation-sensitivity signal identifies predictions that are confident and locally stable under classifier perturbations, while suppressing brittle predictions with high stochastic disagreement. MIRA then performs adaptation with an MI-confidence weighted entropy loss, which down-weights perturbation-sensitive or low-confidence samples before they drive entropy minimization. This weighted loss reduces the influence of brittle predictions that can otherwise cause harmful adaptation under distribution shift. The method requires no source data, retraining, model ensembles, or offline posterior fitting, and can be applied as a plug-and-play modification to entropy-minimization TTA. Experiments on distribution-shift benchmarks show improvements in accuracy, calibration, and negative log-likelihood, suggesting that mutual-information-guided perturbation probing and weighted adaptation yield more robust online TTA than deterministic confidence.

0.24

From Single-Agent Self-Discovery to Multi-Agent Adaptation: Scaling Laws of Partner Feedback and Convention Rigidity in Meta Reinforcement Learning Algorithms

cs.MA ·2026-11-01 ·Haomin Luo, Ersha Li
Abstract · 71 pages, 8 figures, 28 tables

A key milestone toward Recursive Self-Improvement (RSI) is replacing hand-crafted optimizers with meta-learned update rules (such as Disco103 and OPEN) that discover how agents learn. Yet when a self-discovered RL model trains with adapting teammates, does co-evolution sustain capability gains or trap agents in brittle conventions? Across 1,800+ runs in continuous collective robotics (VMAS) and discrete coordination (Overcooked), we establish five laws of multi-agent adaptation. First, four-cell counterfactual partner swaps expose an evaluation illusion: apparent team progress often masks focal regression (+1.066 team gain concealing -0.119 focal loss) and inflates transfer by +0.649 to +1.210. Second, a 288-run physical-coupling sweep ($α\in [0, 1.5]$) reveals a linear scaling law ($β= +0.255$, $p = 0.000488$, Spearman $ρ= 1.0$): live reciprocal feedback keeps Disco103 invariant (AUC 0.297--0.322), whereas severing feedback triggers a monotonic collapse (-0.172 to -0.490), with OPEN establishing a sensitivity spectrum (Disco103 > OPEN > MAPPO). Third, extending navigation training eightfold (1.05M steps) enables Disco103 to overtake MAPPO (+0.087 team return, +0.133 bank gain), while decoupled training boosts cross-partner completion by +10.16% (56.25% vs. 46.09%). Fourth, in Overcooked, MAPPO's self-play conventions induce a 28.13% cross-partner collision rate, whereas Disco103 cuts collisions to 12.77% and achieves 3.33x higher bank return (9.96 vs. 2.99; net gain +6.43), further improved by decoupled meta-adaptation (+1.39 with scripted partners). Finally, perturbation pulses ($p=0.00098$), auxiliary entropy diagnostics (8.485 nats), and 48 multi-lifetime meta-evolved rules ($L_2 \in [2.11, 4.91]$) trace these dynamics to partner-gradient blindness and auxiliary drift, informing multi-agent self-discoverd algorithms.

0.24

SWE-Trail: History-Guided Trajectory Collection for Coding Agent Training

cs.AI cs.SE ·2026-11-01 ·Ziming Li, Lu Yu, Jiatan Huang, Qing Cui, Jun Zhou +1 more
positioncommunity pick
Abstract · Coding agents, Trajectory collection, Agent training data, History-guided exploration, SWE-bench

Coding agents solve repository issues through long sequences of exploration, editing, and validation. These trajectories are increasingly used as post-training data, so how they are collected shapes what the trained model learns. A common approach is to sample multiple trajectories independently for each issue and then select or curate them. However, as more trajectories are sampled independently, later trajectories become increasingly redundant, even though earlier trajectories could guide where subsequent collection effort should be spent. A natural way to use this history is to expose it to the coding agent as advice. But training on such trajectories can make the model rely on advice that is unavailable at test time, creating a train-test mismatch. Instead, we argue that history should guide where collection effort is spent, while the coding agent decides what to generate. We introduce SWE-Trail, a trajectory collector that uses earlier trajectories to guide subsequent collection while keeping that history outside the coding agent's context. During a new rollout, SWE-Trail uses history to identify a state where additional collection effort may be useful. There, the coding agent proposes candidate next actions from its ordinary context, and SWE-Trail briefly executes them from the same checkpoint to decide which route continues. We evaluate collection approaches by training student models on the trajectories they produce. For a 124B student initialized from Ling-3.0-flash-base, SWE-Trail improves mean resolution over independent sampling by 2.33 and 1.60 percentage points on SWE-bench Multilingual and SWE-bench Pro, respectively. Without history, the same pipeline produces more resolved trajectories than independent sampling, but improves by only 0.11 points on Multilingual. These results show the value of using history to guide collection without exposing it to the coding agent.

0.24

State-Aware Virtual Patient Simulation for Training Psychiatrists in Challenging Clinical Scenarios

cs.AI ·2026-11-01 ·Harsh Mohta, C Chandrahas Reddy, Sunhith Reddy Bojja, Manik Inder Singh Sethi, Vinesh Gupta +1 more
Abstract · work in progress

Psychiatrists must be prepared to manage patients exhibiting challenging psychological states, including suicidal ideation, hallucinations, and other high-risk behaviors. However, opportunities to safely practice such interactions are limited. While recent conversational agents can simulate patients, existing approaches largely rely on large language models (LLMs) to determine patient responses, making it difficult to consistently maintain psychological state, adapt to the clinician's responses, and generate clinically plausible challenges that meaningfully test the clinician. We present a state-aware virtual patient simulation pipeline for training psychiatrists in challenging clinical scenarios. The proposed system maintains an evolving patient state throughout the interaction and orchestrates patient behavior rather than leaving response generation entirely to an LLM. At each interaction, a learning-based behavioral strategy selector determines the type and level of challenge appropriate to the current patient and clinician context. The selected strategy, patient state, clinician response, interaction history, and relevant clinical knowledge are then provided as grounded context to an LLM to generate a semantically relevant and psychologically consistent patient response. Following each session, the system analyzes the interaction and provides feedback to the clinician on their communication and management of the virtual patient.

0.23

State-Augmented Learning for Post-Deployment Adaptation of Coexisting O-RAN Applications

cs.LG cs.NI ·2026-11-01 ·Wonhyeok Ko, Yeonseo Jeong, Songnam Hong
Abstract · submitted to IEEE Transactions on Mobile Computing

Fixed control policies in Open RAN can require retraining when service requirements, traffic loads, or interacting applications change. We propose ATTUNE, a state-augmented learning method for an energy saving application that coexists with a learned mobility load balancing application under a service constraint. ATTUNE conditions the energy saving policy on a Lagrange multiplier and updates this scalar during operation using measured service degradation, while keeping both applications' network parameters fixed. To train a single policy across service requirements, ATTUNE samples target service levels uniformly and maps them to multipliers through an adaptive table. This procedure accounts for policy-dependent multiplier responses and improves energy savings compared with training over a widened fixed multiplier range. We characterize the induced distribution of service levels and the convergence and fluctuations of the online update. An exact finite-window identity establishes convergence of average measured degradation to the target when the projection corrections vanish on average. Simulations show that a single ATTUNE policy achieves energy savings comparable to policies trained separately for each service requirement. It adapts to changes in the target, load, and partner application without retraining, increases energy savings at light load, and maintains target tracking for separately trained application pairs. No statistically significant increase in cell switching is detected in the evaluated comparisons.

0.23

When Should an Agent Refresh? Adaptive Index Reuse for Efficient Long-Horizon Reasoning

cs.LG cs.CL ·2026-11-01 ·Guangyuan Wang, Yifan Zhang, Shilong Liu
benchmark
Abstract

Language agents read code, tool output, and dialogue in contexts that grow at every step. Dynamic sparse attention lets a small indexer choose the earlier tokens each query attends to, but the indexer still scores the whole prefix at every layer. Cross-layer reuse methods such as IndexCache rerun the indexer at only some layers, with one fixed pattern for every input. In an agent, the relevant evidence can change abruptly when a traceback or test result arrives, so fresh selections matter more at some steps than at others. We propose Adaptive Index Reuse (AIR), which keeps IndexCache's execution and learns, for each segment of an agent episode, how many layers refresh their selection. A small controller reads signals from completed computation and the latest tool event, chooses a pattern from a nested library, and learns from terminal task reward under an expected compute budget. On Qwen3-4B-Instruct-2507 with a 128K context, AIR improves success over the strongest static schedule at matched cost on SWE-bench Verified (27.0% to 32.4%), Terminal-Bench 2.0, and three $τ^2$-bench domains, and it lowers time per successful task on all five. Across 50 backbone, context, and benchmark settings, it matches or exceeds that schedule in 44.

0.23

Xiaomi Auto World Model 2.0: Causal World Models for Real-World Driving and Embodied Intelligence

cs.AI cs.LG ·2026-11-01 ·Zhenxin Zhu, Zhijian Huang, Cheng Chi, Yuru Wang, Jianhua Wu +18 more
code/data
Abstract · 40 pages, 20 figures. Project page: https://xiaomiautoworldmodel.github.io/CausalWorldModel/

We present Xiaomi Auto World Model 2.0, a unified streaming causal world model for Physical AI. It models and generates causal physical evolution in the real world, and unifies heterogeneous conditioning inputs for view-consistent generation. The architecture integrates View-Temporal Attention (VTA) to capture cross-view spatial dependencies and temporal dynamics, together with Multi-view Rotary Position Encoding (RoPE) to encode positional relationships across views and time. WorldRec 2.0 reconstructs static and dynamic 3D Gaussians from multi-view, multi-temporal observations in a single forward pass and uses their renderings as explicit scene context to preserve geometry and appearance during long-horizon generation. Conditioning inputs are fused through VAE latent concatenation in a shared representation space. We incorporate Gated Linear Attention (GLA) into the temporal layers of VTA for efficient streaming inference, reducing computation and KV-cache memory during multi-view rollouts. A physical-interaction data pipeline covers diverse interactions and long-tailed events. From Pre-train onward, these data are combined with large-scale videos and heterogeneous visual modalities to learn visual, geometric, and dynamic priors. During Mid-train, training across diverse view configurations further improves the model's ability to capture how actions and object states shape subsequent scene evolution. After distillation-based Post-train, we apply the model to closed-loop simulation, closed-loop training, and World Action Model (WAM) applications. These applications generate physically consistent future states and support world-state prediction, policy learning, and action generation for Physical AI.

0.23

Uncertainty-Aware Joint Generative Modeling of Fault Event Sequences and Sensor Time Series in Solar PV Plants

cs.LG eess.SY ·2026-11-01 ·Jiayu Huang, Boyang Xu, Yongming Liu, Hao Yan
Abstract · Author preprint of a paper accepted to IEEE Transactions on Automation Science and Engineering; 18 pages including supplementary appendix

This paper presents a multimodal generative model designed to simulate a solar energy plant system for predictive maintenance and fault prediction. We propose a novel approach to enhance system generation through conditional generation by leveraging both event sequences and time-series data and improve next event prediction. The framework employs a Transformer Hawkes Process (THP) to encode event data and an iTransformer to encode time-series data, integrated via a co-attention mechanism that effectively captures cross-modal dependencies between discrete events and continuous signals. From these integrated representations, we employ two specialized decoders for simulation: a Conditional Latent Diffusion Model (CLDM) for modeling future sensor trajectories and the Score Matching-based UnceRtainty quantiFication for Transformer Hawkes Process (SMURF-THP) for high-fidelity event prediction. This dual-decoder strategy supports uncertainty-aware, history-conditioned simulation by rolling the model forward for a specified number of generation iterations, producing finite stochastic scenarios rather than deterministic extrapolations over the full observational record. Evaluated on a comprehensive dataset from the Red Rock Solar Site in Arizona, our approach improves aggregate event-prediction F1 relative to baselines and supports applications in predictive maintenance, fault detection, and energy optimization for solar power systems.

0.23

Order-Only On-Policy Distillation

cs.LG cs.CL ·2026-11-01 ·Guangzeng Han, Yu Wu, Xiaolei Huang
community pick
Abstract

On-policy distillation (OPD) trains a student on its own generations by matching the teacher's token probabilities. These probabilities entangle two kinds of information: which tokens the teacher prefers, and how confident it is. The latter reflects the teacher's own capacity and calibration, which a smaller student may be unable to reproduce. Because the teacher temperature changes these values without changing the ordering, OPD is also sensitive to the temperature choice. We ask whether the teacher's ordering of candidate tokens is enough, and propose Order-Only On-Policy Distillation (O$^{3}$PD). At each student-generated prefix, O$^{3}$PD trains the student to reproduce the order of the teacher's top-$K$ tokens, without fitting their probability values. The objective is therefore invariant to the teacher temperature and needs no temperature tuning. Across six teacher--student pairs, O$^{3}$PD outperforms OPD on the average over HMMT25, AIME24, and AIME25 in every pair, by up to 2.9 points, with gains of up to 5.6 points on AIME25. Other ranking losses yield similar gains, so the benefit comes from ordinal supervision itself. Analysis shows that O$^{3}$PD departs further from the teacher's probabilities than OPD, yet reproduces its ranking more faithfully and yields a stronger student. The ranking, not the probability values, thus carries the knowledge that on-policy distillation needs to transfer.

0.23

Concise-9B: Removing the Slack from a Reasoning Model's Thinking

cs.LG cs.CL ·2026-11-01 ·Alexander Boesgaard Lorup
Abstract · 18 pages, 5 figures, 14 tables

Reasoning models answer better when they think longer, but a thinking trace keeps going well after its answer has formed. We measure this slack in Qwen3.5-9B and remove most of it with concise self-training: two rounds of on-policy rejection-sampling fine-tuning that keep the shortest correct samples, plus a small share of answer-synthesis targets, giving Concise-9B. On 8,983 held-out problems (competition and hard mathematics, SuperGPQA, a five-area confirmation set and MMLU-Pro), Concise-9B generates 0.66x the tokens of Qwen3.5-9B at a 32K thinking budget, with accuracy equivalent within +/-1 point (-0.20 [-0.77, +0.36] points). Forcing each trace on the hard and confirmation sets to answer every 2,048 thinking tokens, we find that 54-64% of Qwen3.5-9B's thinking tokens come after the point from which every forced answer equals the final one (41-48% for Concise-9B). Stopping there would keep every answer at 0.35-0.44x the base model's tokens, but finding that point takes hindsight: stopping when three consecutive forced answers agree costs the base model 2.4-9.2 points. Concise self-training removes slack without that cost: 86% [83, 90] of the thinking tokens it saves come after the answer has formed. It has a limit: with traces continued to 80K tokens, Concise-9B solves 25 fewer of 1,276 hard problems (-25 [-49, -1]), mostly in competition mathematics. The measurement also yields a cheap commit signal for adaptive sampling that matched two-sample escalation within +/-1 point at 0.67x its tokens in a pre-registered test. Results on Qwen3.5-4B, on which the recipe was developed, and a replication on Olmo-3-7B-Think suggest that the saving depends on how verbose the base model is.

0.22

Wave Physics as Representation: Tensor-Structured Neural Reconstruction for Ocean Acoustic Fields

cs.LG cs.AIcs.SDeess.AS ·2026-11-01 ·Yifan Sun, Panqi Chen, Weifeng Shi, Xiao Fu, Lei Cheng
Abstract

Reconstructing three-dimensional (3D) complex ocean acoustic fields from sparse measurements is challenging due to highly oscillatory wave structures and restricted sensing patterns/routes. Learning-based approaches fall into two categories. First, purely data-driven models can learn complex field patterns, but do not explicitly exploit wave physics that may enhance interpretability and reduce training costs. Second, physics-informed methods impose a specified physical equation as a reconstruction constraint, but often require environment-dependent parameters that are difficult to acquire. As a result, practical 3D acoustic field reconstruction that is both physically grounded and robust to uncertain environmental information remains largely unexplored. This work takes an alternative route: instead of solving a fully specified wave equation, it extracts a structured representation from wave physics and learns its components from data. From the Helmholtz equation and normal mode propagation, we derive a structured spatial-angular tensor representation, termed the S/c representation. Under a low-rank specialization, it takes the form of a rank-(L,L,1) block-term tensor decomposition, which connects structured sampling patterns/routes used in ocean sensing to sufficient conditions for field recovery. Building on this tensor representation, we propose TIDAL, a neural reconstructor that uses the S/c representation as an inductive bias for structured decoding and aligns information aggregation with the sampling geometry. The neural learner removes the need for known model parameters required by conventional physics-based recovery methods. Experiments across three structured sampling geometries show that TIDAL achieves the best reconstruction performance and fastest inference among baselines, with benefits from the physics-derived representation under limited data and model capacity.

0.22

Renormalized Cellular Intelligence: When Does Model Reduction Preserve Collective Capability?

cs.MA ·2026-11-01 ·Xuening Wu, Yanlan Kang, Qianya Xu, Haode Yang, Shan Yu +1 more
Abstract

As multi-agent systems grow in size and complexity, understanding which features of their interactions support collective capability becomes increasingly difficult. Can a simpler description preserve both task performance and the measured contribution of interactions between agents? Preserving collective dynamics alone is insufficient: reduction can alter decision-relevant information or give a non-communicating baseline access to pooled observations, distorting the measured advantage of interaction. We introduce Renormalized Cellular Intelligence (RCI), a framework for specifying and auditing capability preservation across scales. RCI treats the task, interacting dynamics, matched baseline, decision interfaces, and resource budget as a joint object of reduction. It specifies dynamical closure, decision compatibility, and preservation of the intended baseline comparison, and bounds the resulting error in relative capability gain. An exact analytical construction demonstrates task-score preservation, while a counterexample shows how closed reduced dynamics can erase the measured interaction gain. Using cellular systems as models of locally interacting agents, we evaluate three reduction candidates on five frozen learned models without retraining. The final candidate reduces the evolving state from eight to five dimensions per cell (37.5%). Within-run uncertainty bounds support relative gain above a prespecified threshold for the fixed model collection, yet the candidate fails the closure and decision-preservation requirements for a complete RCI certificate. RCI provides explicit criteria for distinguishing positive interaction gains from

0.22

HyVox: Rethinking Low-Power Voice Intelligence towards Neuromorphic-von Neumann Edge Computing

cs.NE cs.AIcs.LGcs.SD ·2026-11-01 ·Snehasis Addy, Eduardo Calle Ortiz, Prashant Shenoy, Phuc Nguyen
Abstract

We present HyVox, a hybrid neuromorphic--von Neumann system for low-energy open-set speaker identification under tight energy budgets. HyVox splits the workload at the speaker embedding. A neuromorphic processor embeds the speech with sparse, event-driven computation, and a von Neumann microcontroller holds the speaker gallery and makes every decision: it scores queries by cosine similarity, rejects unknown speakers, and enrolls new users by writing a prototype, with no weight update. On the neuromorphic processor, a hierarchical temporal abstraction (HTA) SNN backbone learns a synaptic time constant per leaky integrate-and-fire (LIF) neuron, allocating 832 neurons across fast, medium, and slow timescales. A 32-neuron spiking embedding extractor (SEE), pretrained on episodes that each enroll new speakers and reject outsiders, turns the backbone output into an embedding for speakers unseen in training; its 32 spike counts are the only interface between the two processors. To fit the chip's synapse and integer budgets, we compress the backbone 6.5x to 29.0k synapses, train it for integer bit-shift dynamics, and fine-tune it with chip-accurate spiking training (CAST). CAST trains through a differentiable replica of the chip's integer neuron update, derived from its cycle-accurate simulator, so that gradient descent optimizes the exact weights and thresholds the chip runs; it doubles the top-1 accuracy of the mapped backbone on unseen speakers. On 20 Common Voice households of unseen speakers, HyVox reaches 99.6\% top-1 and 5.5\% rejection EER with 2 members and 94.9\% and 13.2\% with 20, and the network deployed on the neuromorphic chip matches its trained model. Each inference on a 3-s window takes 0.96 s and 1.63 mJ of dynamic energy, 42\% lower latency than the fastest embedded ANN baseline and 13.5x less dynamic and 5.8x less total energy than the most energy-efficient one.

0.22

Atom-Wise Allocation of Message-Passing Scale for Molecular Property Prediction

cs.LG ·2026-11-01 ·Regina Qianru Zhang, Honggang Wen, Zongru Li, Jinsong Ouyang, Zhiqin Chu +3 more
Abstract

Molecular properties depend on chemical structure at different ranges, from local polar groups governing solvation to extended substructures shaping bioactivity, yet graph neural networks usually fix one propagation scale for every atom. We present PRISM (Per-atom Routing over Intermediate States of Message Passing), in which each atom's encoded features produce a bounded scalar that positions a unimodal distribution over retained message-passing states and matching circular-fingerprint radii. Because one coordinate translates the whole profile, allocated depth responds to the routing scalar in closed form: at fixed temperature it is nondecreasing, its sensitivity scales with the profile variance, and it saturates in the clipped regions. Beyond providing closed-form analyzability, this constrained one-parameter profile matches or significantly outperforms parameter-matched, unconstrained node-wise gates and Jumping Knowledge attention. Under scaffold partitions with equal tuning budgets against eight graph, fingerprint, and geometric baselines, PRISM trained from scratch without three-dimensional coordinates achieves the lowest mean error on aqueous solubility and hydration free energy and strong accuracy on BACE and BBBP. As a general layer-combination rule, PRISM transfers directly to GCN, GIN, and GAT backbones, reducing hydration free energy error by up to 29% relative to last-layer readout. Applied to a pre-trained GIN, it improves on last-layer readout on four of six endpoints and gives the lowest lipophilicity error overall. In trained models, allocated depth differs systematically between atom types in a task-dependent way, and permuting profiles among a molecule's atoms raises prediction error on several endpoints, so the rule acts as a learned counterpart of atom typing.

0.22

Optimization Frameworks & Learning Algorithms for Discrete-Time Mean Field Games

cs.LG math.OC ·2026-11-01 ·Xin Guo, Mathieu Laurière
Abstract

This monograph introduces discrete-time mean field games with finite state and action spaces, emphasizing optimization formulations and learning algorithms. Starting from Markov decision processes and finite-player stochastic games, it develops the notion of mean field Nash equilibrium, establishes existence and uniqueness results under suitable assumptions, and explains how mean field equilibria yield approximate equilibria in large finite-player games. Building on the linear programming formulation of Markov decision processes, occupation measures and primal-dual formulations connect equilibrium computation with mathematical optimization. The monograph presents fixed-point iterations, fictitious play, online mirror descent, and optimization-based methods, discussing their convergence properties and limitations. Numerical examples illustrate crowd motion, traffic routing, epidemic management, and the computation of multiple equilibria. It combines examples, theoretical results, and implementable algorithms, with bibliographic notes providing connections to the broader literature.

0.22

Who Should Score the Cache? Mid-Depth Layers Are the Best Judges of KV-Cache Eviction

cs.LG cs.CL ·2026-11-01 ·Ekaterina Seregina
Abstract · 19 pages, 3 figures

Key-value (KV) cache eviction methods rank tokens by attention scores, and each layer scores its own cache. To our knowledge, no eviction method asks which layer should score. We find, across twenty-one models, that the depth of the scoring layer governs eviction quality. On the eighteen models with per-layer caches, spanning 0.5B to 109B, long-context retrieval accuracy collapses at shallow layers, peaks between 0.4 and 0.75 of depth, and degrades at the last layer. The curve arises because a layer judges tokens well only once it has used the question to find the answer, which takes several layers. Modeling eviction as a top-$k$ cut, we compute answer survival probability from prefill attention and reproduce the measured curves. This approach selects the scoring layer using calibration prompts with known answers, and the selected layer is the best layer measured on most models and never significantly below it. It improves on the own-layer default on sixteen of the eighteen models, with significant gains on seven, all at 14B or below, while on larger models the default is already close to the best mid-depth layer. Scoring every cache from the selected layer while keeping one set of tokens for all heads also beats per-head SnapKV by 18 to 43 points on five models up to 14B under tight budgets. This gain carries over to financial filings at 32K context, all at the same memory budget, with no extra forward pass. The same curve explains a penalty in shared-cache models, where several layers read one cache and the layer that writes it scores by default. On Gemma-4-E2B the writer is shallow and loses 28 points to a mid-depth reader at a 10% budget. We also find that in open-book question answering (LongBench) no mid-depth layer beats the own-layer default.

0.21

Sharp Generalization for Nonparametric Regression in Interpolation Space by Shallow Over-Parameterized Neural Networks with Channel Attention

cs.LG stat.ML ·2026-11-01 ·Yingzhen Yang, Ping Li
theory
Abstract

We study nonparametric regression using an over-parameterized two-layer ReLU neural network with fixed covariance-based channel attention, trained by gradient descent with both layers updated simultaneously. The spherical covariates follow an arbitrary fixed non-atomic distribution, and the target function lies in an interpolation space with spectral bias widely studied in the statistical learning literature. We prove that early-stopped training achieves regression risk $\cO(\eps_n^2)$, where $\eps_n$ is the critical population rate of the attention kernel $\Kattn$. Its positive eigenvalues are $λ_j^{s'}$, where $λ_j$ are those of the ReLU feature covariance kernel $K$ and $s'=2s+3$ for $s\ge1$. Under polynomial eigenvalue decay $λ_j\asymp j^{-2α}$ with $α>1/2$, the regression risk bound $\cO(\eps_n^2)$ specializes to the minimax optimal rate $\cO(n^{-2αs'/(2αs'+1)})$. Our analysis is based on two new technical results on the optimization and generalization of the attention network. First, the neural network function after every gradient descent iteration admits a decomposition into a function with bounded Reproducing Kernel Hilbert Space (RKHS) norm and a uniformly small residual. Second, local Rademacher complexity converts this decomposition into a regression risk bound. Kernel gradient descent on the same ball attains the same rate, while our theorem gives explicit sufficient width and auxiliary-sample conditions for simultaneous network training. Our analysis is further extended to covariates supported on $\RR^d$ through a novel normalization method. Under a sub-Gaussian covariate distribution, the attention network achieves a nearly optimal regression risk $\cO((\log n)n^{-s'(d+3)/(s'(d+3)+d)})$ without a density assumption.

0.21

Overcoming Demand Censoring through Randomization for Risk-Sensitive Newsvendor Learning

cs.LG math.OC ·2026-11-01 ·Zhengchao Wang, Nam Ho-Nguyen
Abstract · 30 pages of main text and 4 figures

We study online newsvendor learning with bounded demand, censored observations, and convex nonlinear shortage costs. Because stockouts can conceal the shortage magnitudes needed by gradient-based algorithms, we use randomized ordering to construct unbiased subgradient estimates. We characterize admissible randomization distributions within this framework and develop an online inventory policy that combines the resulting subgradient estimates with gradient-based updates. Our regret analysis quantifies the cost of randomization via the estimation variance and an additional loss term. We characterize the randomization distributions that minimize this objective. For piecewise-linear shortage costs, the optimal distribution is computed through a linear-time recursion; for costs with bounded curvature density, it is characterized by a scalar ordinary differential equation. The policy achieves expected regret of $O(T^{2/3})$ against the best fixed order in hindsight over $T$ periods. Under independent demand and suitable curvature of expected losses, the bound improves to $O(\sqrt{T})$. Matching minimax lower bounds establish rate optimality over the corresponding classes with a fixed marginal-cost bound. Both rates extend to nonperishable inventory under uniformly positive conditional expected demand. Numerical experiments on synthetic and retail data show improvements over alternative policies using censored observations.

0.21

DyRadio: Spatiotemporal Diffusion for Dynamic Radio Map Reconstruction

cs.LG cs.AI ·2026-11-01 ·Zhijie Fan, Zhifeng Zhao, Rongpeng Li
Abstract · 13 pages, 7 figures, 8 tables

Accurate radio map (RM) reconstruction is essential for environment-aware wireless networks. Sampling-based methods, which infer radio conditions from sparse measurements, are widely adopted because they do not require explicit knowledge of transmitter configurations. However, in dynamic environments, time-varying propagation conditions cause RMs to evolve, complicating reconstruction from temporally distributed sparse observations. Although cross-frame observations provide complementary spatial information, their direct aggregation can introduce temporal state mismatch. Furthermore, unknown measurement noise adds to the reconstruction difficulties. To address these challenges, we propose DyRadio, a framework that integrates spatiotemporal RM reconstruction with unknown noise estimation. Specifically, a propagation-prior-guided, noise-conditioned spatiotemporal diffusion model (DM) jointly reconstructs consecutive RMs by capturing cross-frame dependencies, while adapting to measurement noise levels. To estimate the noise level, an iterative predictive likelihood method alternates between DM-based blind-area prediction and noise-level updates. The estimated noise level then conditions reconstruction of the complete RM sequence. Experimental results demonstrate superior reconstruction accuracy and robustness against measurement noise, with peak signal-to-noise ratio (PSNR) improvements of 1.18-5.75 dB over the best-performing baseline across all evaluated settings.

0.21

Same Function, Different Gradients: Evaluating Gradient Reliability under Reparameterization

cs.LG ·2026-11-01 ·Bum Jun Kim
Abstract · 47 pages, 7 figures

Gradients are no longer used only to train models: Gradient magnitudes are now measurement instruments, used to rank attention heads for pruning, to decide per-example clipping in private training, to allocate adapter rank budgets, and to debug models. Across these workflows, larger gradient magnitudes are taken to mean greater importance. Even for a fixed input--output function, these gradients are computed in one of many possible parameter coordinate systems. For example, attention scores depend on the query and key projections only through the paired product $W_QW_K^\top$. Scaling a head's query projection by $c$ and the corresponding key projection by $1/c$ therefore leaves every output logit unchanged. This change in parameter coordinates rescales the corresponding query and key gradient norms by $1/c$ and $c$, respectively. To test whether gradient-based decisions remain unchanged under these coordinate changes, we introduce Gradient Reliability under Reparameterization (GRR). GRR rewrites a trained model in new parameter coordinates that compute exactly the same function, verifies that predictions are unchanged, and then tests whether the decisions derived from the gradients are unchanged too. Our audits span GLUE classifiers, a ViT on CIFAR-10, GPT-2-medium and GPT-2-large, a 1.1B-parameter Llama-style model on WikiText-2, and six trained LoRA adapters. Empirically, predictions are preserved throughout, but both the gradients and the gradient-based decisions change. Across 66 GLUE attention audits, the median top-3 overlap of raw head-gradient rankings is only $1/3$, and rankings and pruning masks change even under the mildest rescalings we test. In the largest of 14 global-pruning audits, the shifted ranking selects head masks that differ in four of 22 pruned heads and reach $50.92\%$ and $94.84\%$ validation accuracy on one ELECTRA SST-2 model.

0.21

ERIA: An Instruction-Optimized Agent for Clinical EEG Reporting

cs.LG cs.AI ·2026-11-01 ·Ahmed Sameh, Gayal Kuruppu, Neeraj Wagh, Bridger Brundy, Yogatheesan Varatharajah
Abstract · 15 pages, 2 figures, 3 tables

Clinical EEG interpretation requires integrating technical quality, background and behavioral states, interictal and ictal abnormalities, and overall clinical context across long recordings. We introduce ERIA (EEG Reporting and Instruction-Optimized Agent), a locally deployable framework in which a single 27B language-model controller analyzes whole-recording EEGs by orchestrating specialized parametric and signal-processing tools. ERIA combines learned tools for background abnormality, sleep-wake state, seizure-like activity, and interictal epileptiform discharges with deterministic spectral, amplitude, symmetry, montage, and signal-quality analyses. Rather than fine-tuning the controller or using multi-agent orchestration, ERIA keeps the controller weights, specialist tools, and safety constraints fixed and optimizes only its natural-language analysis policy with GEPA (Genetic-Pareto), using feedback from expert-written EEG reports. GEPA improves physician-aligned whole-recording report quality over the raw policy, and the optimized policy transfers to external EEG benchmarks. On CerebraGloss-Bench and BrainBench Foundational Analysis, ERIA remains competitive with the models evaluated in the original benchmarks. These results show that specialist tools and language-level policy optimization can extend the clinical and general EEG capabilities of a single locally executable model while preserving local inference and traceable tool use.

0.21

Learning by Doing: Autonomously Compiling and Curating Agent Skills from Verified Traces

cs.AI cs.CL ·2026-11-01 ·Daniel Chalef, Peter Evans, Pavlo Paliychuk, Mehul Patel, Preston Rasmussen +1 more
Abstract · 25 pages, 9 figures, 14 tables

Memory systems for production LLM agents mostly store semantic memory (facts about users and domains) or episodic memory (earlier interactions). Research systems such as Trace2Skill, WikiSkill, and SkillRevise learn procedures from agent runs, package them as Skills, and evaluate them on benchmarks. We describe Zep Agent Skills, a production service that gives deployed agents this procedural memory. The service uses only verified attempts as evidence, cites the supporting events for each field of a Skill, and governs the admission and distribution of each Skill version. The service records each task attempt as a Trace with ordered events, an outcome, and a verification strength. A compiler with separate Reflector and Curator roles turns verified Traces into versioned Skills. Scans and admission gates control which candidates become active, and verified use outcomes change the search rank of each Skill. The Agent Skills format is an open standard that Anthropic originated and donated to the Agentic AI Foundation, so any host of the format can load a compiled Skill. Agents retrieve Skills through lexical search with progressive disclosure, over an API or MCP. We evaluate Zep Agent Skills on $τ^3$-bench banking, AppWorld, and EnterpriseOps-Gym. Skills compiled from the Traces of a teacher model increase the pass rate of three weaker executor models on the same tasks, at every reasoning effort, by 4.5 to 18.8 points. Most of the gain is on the tasks that the teacher solved. On ALFWorld, Skills compiled from the runs of a small model give a gain similar to or larger than the gain that prior work reports for continuous fine-tuning, with no training.

0.20

When Does Sparse Routing Help? The Geometry of Expert Separation in Mixture of Experts

cs.LG stat.ML ·2026-11-01 ·Huy Nguyen, Dung Le, Alessandro Rinaldo, Nhat Ho
Abstract · 93 pages

Sparse mixtures of experts combine multiple prediction models while evaluating only a small subset for each input. Although this architecture reduces computation, how expert separation and routing sparsity jointly govern expert recovery remains an open problem. In this paper, we address this question for location Gaussian mixtures of experts with a fixed number of experts by establishing how expert separation and spatial arrangement determine recovery rates. For dense models, our results reveal that when experts form well-separated clusters, the separation geometry is governed by the size of the largest cluster; without a prescribed cluster structure, this geometry is instead driven by the total number of experts. For sparse models, the separation geometry on retained routing regions is determined by the number of active experts, while the contribution of omitted regions is controlled by their total probability. Consequently, activating fewer experts reduces the separation complexity in these bounds. Furthermore, Top-1 routing yields a particularly striking result: experts can be recovered at a rate independent of the separation quantity. Our technical approach combines polynomial interpolation and Gaussian reproducing-kernel Hilbert space (RKHS) test functions with quantitative restriction inequalities that extend pointwise comparisons to expert functions.

0.20

MatMoE: Matryoshka Mixture-of-Experts with Dynamic Mixed-Precision Quantization for Efficient Inference

cs.LG cs.AI ·2026-11-01 ·Renjie Wei, Haochen Huang, Chuyu Qiu, Jinqi Wen, Dehao Xu +1 more
published
Abstract · Accepeted by ICCAD 2026

Mixture-of-Experts (MoE) architecture has been widely adopted in large language models (LLMs) since it reduces the computational cost through sparse expert activation. However, the massive parameter scale of MoE models hinders their deployment on resource-constrained edge devices. Although existing mixed-precision quantization reduces the model size, they are inherently static, fixing the bit-width of each expert across decoding steps. Such static schemes yield suboptimal performance as they overlook the dynamic variation in expert importance during decoding. Furthermore, they fail to adapt to the dynamic latency constraint driven by the fluctuating system utilization. To address these challenges, we propose MatMoE, a dynamic mixed-precision quantization framework for efficient MoE inference. MatMoE features: (1) a Matryoshka MoE PTQ algorithm that enables experts to operate flexibly at multiple precisions within a stringent memory budget; (2) a floor-and-compensation quantization method that eliminates redundant memory accesses in the Matryoshka representation; and (3) a dynamic bit-width allocation strategy that determines the precision for each expert based on their routing scores to satisfy the dynamic latency requirement. We implement our MatMoE system atop llama.cpp for both CPU and GPU platforms. Extensive evaluations across three modern MoE models and challenging benchmarks demonstrate that MatMoE achieves $4.0\% \sim 18.3\%$ higher accuracy than state-of-the-art static mixed-precision method. Compared to llama.cpp, MatMoE delivers $1.3\times \sim 3.7\times$ speedup on Raspberry Pi 5, and $1.2\times \sim 1.3\times$ speedup on NVIDIA RTX 4090 and Jetson AGX Orin.

0.20

Evolving Memory, Learning to Meta-Control: Test-Time Adaptation for Experience-augmented LLM Agents

cs.LG cs.AIcs.CL ·2026-11-01 ·Zeyuan Zhao, Yuanyuan Shi, Peng Yu, Lingyue Fu, Shao Zhang +1 more
Abstract · 23 pages, 5 figures. Submitted to ACL Rolling Review (October 2026)

Self-evolving agents accumulate experience from task streams, but existing frameworks typically rely on fixed pipelines to retrieve and utilize this experience, neglecting when and how to deploy it. We introduce Online Meta-Control of Experience (OMCE), a framework that updates experience memory and learns a meta-control policy from task feedback during test-time adaptation. OMCE employs a lightweight contextual-bandit meta-controller that uses compact retrieval signals to select among four cognitive actions: DIRECT, EPISODE, GUIDE, and DISTILL. This routing mechanism operationalizes three coupled decisions: whether to invoke experience, which memory format to utilize, and when to consolidate new guidance. OMCE evolves online, leveraging task feedback to update the memory bank and refine the routing policy via contextual bandit updates. Experiments on four online adaptation benchmarks across four backbones show that OMCE improves first-attempt success by up to 20.5 percentage points over the strongest compared baseline, while reducing per-task token use by a factor of 5.7 to 62.5 relative to ACE with GPT-5.2. Controlled ablations further support the benefits of learned routing and continued routing policy adaptation.

0.20

FedUDG: Federated Domain Generalization with Unlabeled Clients

cs.LG cs.AI ·2026-11-01 ·Junyu Ren, Wensheng Gan, Philip S. Yu
code/data
Abstract · 12 pages, 4 figures

Federated domain generalization (FDG) aims to learn models that generalize reliably to out-of-distribution (OOD) environments without centralizing client's data. However, existing FDG methods assume that all source clients have labeled data, limiting their ability to leverage the abundant unlabeled client data available in practice. To address this limitation, we propose FedUDG. At the federation level, FedUDG utilizes a parameter-functional differentiation strategy that maintains a semantic anchor updated solely using ground-truth labels. The remaining parameters absorb knowledge from unlabeled domains under bounded aggregation, preventing domain-biased noise from eroding the class semantics. Within each unlabeled client, FedUDG utilizes a student-teacher framework with two complementary objectives: (1) semantic self-training on high-confidence samples, using the functional differentiation between the semantic anchor and classifier to constrain the model to learn semantically pure and cross-domain stable features; and (2) label-free representation distillation on all unlabeled samples to stabilize learning under appearance variations. Together, these objectives guide the model toward a transferable feature space, while mitigating overfitting to domain-specific characteristics. Furthermore, to address the lack of systematic evaluation of FDG under formal privacy constraints, we develop a sample-level differentially private extension, DP-FedUDG, to quantify the trade-off between privacy and OOD generalization. Across four evaluation protocols, FedUDG consistently outperforms state-of-the-art baselines. The code is available at https://github.com/Jr10426/FedUDG

0.19

Beyond Validation Loss: Coverage and Reliability Dynamics in Robot Policy Pretraining

cs.LG cs.RO ·2026-11-01 ·Zhanyi Sun, Max Simchowitz, Shuran Song
code/data
Abstract

In modern robot learning, generative robot policies are first pretrained through supervised learning from demonstrations, then refined through reinforcement learning, supervised finetuning, or other forms of posttraining. Yet conventional pretraining metrics, such as validation loss and average task success, provide an incomplete account of which policies support both effective direct deployment and subsequent posttraining. In this work, we study closed-loop pass@k, which evaluates the probability that k rollouts from the same initial condition attain at least one success. Thus, pass@1 measures mean task success; large-k pass@k measures whether success remains reachable under repeated sampling. Across a wide range of posttraining recipes, we show that large-k pass@k reveals posttraining differences not captured by pass@1. In particular, checkpoints with similar pass@1 but different large-k pass@k have different posttraining performance. Moreover, we show pass@k peaks at different checkpoints during supervised pretraining, revealing a coverage-reliability tradeoff over the course of training. Based on these observations, we propose two practical interventions: (1) a purely offline checkpoint-selection criterion that balances coverage and reliability without closed-loop evaluation, and (2) a generative data augmentation procedure that expands supervision beyond the finite demonstration set, preserving large-k pass@k while maintaining or improving pass@1 and raising the observed coverage-reliability Pareto frontier. Together, our results motivate evaluation, checkpoint selection, and training procedures that explicitly account for posttraining. Blog post, visualizations, and code are available at https://zhanyisun.github.io/robot-policy-learning-dynamics/.

0.19

Reinforcement Learning in High-Energy Physics: A Survey from Accelerator Control to Theory Exploration

cs.LG hep-exhep-ph ·2026-11-01 ·Luca Anzalone
survey
Abstract · 86 pages, 5 figures, 3 tables

Reinforcement learning (RL) is a machine learning paradigm in which an agent learns a decision-making policy by interacting with an environment, guided by a scalar reward rather than explicit supervision. Supervised and unsupervised learning have become indispensable tools in high-energy physics (HEP), but they learn from fixed datasets, whereas many HEP tasks are closed-loop decisions: the data an algorithm observes depend on its choices, the feedback is sparse or delayed, and the objective is often measured on a real machine or in a stochastic simulation, not by a differentiable model. This survey reviews the application of RL to such tasks: accelerator control and feedback, data quality monitoring, jet grooming, clustering, and representation, charged particle tracking, beyond-the-Standard-Model (BSM) and string model building, detector design, trigger control, analysis, and formal quantum field theory computation. It provides a self-contained introduction to the underlying RL methods, from model-free and model-based to multi-agent, offline, inverse, and bandit approaches, reviews each domain against its traditional methods, and summarizes the head-to-head comparisons. The applications range from policies deployed on real accelerators and evaluated on recorded collision data to exploratory studies in formal theory. RL is most advantageous for repeated tasks with a fast simulator, decisions inside fast control loops, operation under faults and corrupted measurements, physics-based rewards in place of labels, and sequential discrete constructions with a reusable policy, while Bayesian optimization, genetic algorithms, and differentiable methods remain competitive elsewhere. We conclude with the open challenges that recur across domains, including reward design, sim-to-real transfer, safety, the lack of shared benchmarks, and the statistical validity of policy-driven decisions.

0.19

SSPE: Robust Sequential Staypoint Extraction in Degraded Trajectory Data

cs.LG cs.DB ·2026-11-01 ·Yueyang Liu, Lance Kennedy, Andreas Züfle
Abstract

Extracting semantically meaningful staypoints from raw trajectory streams is a foundational prerequisite for mobility analytics. However, real-world trajectory data is frequently degraded by severe spatial noise and protracted temporal dropouts. Classic threshold-based extraction methods, such as the widely adopted sliding-window algorithm, are highly sensitive to these anomalies, routinely fracturing continuous stationary episodes into disjointed fragments. To overcome this limitation, we propose the Sequential Staypoint Extraction (SSPE) algorithm. SSPE employs a resilient, state-driven architecture that replaces static spatial anchors with a dynamic centroid and introduces an $n$-step lookahead heuristic to structurally tolerate transient coordinate drift. Concurrently, it utilizes temporal bridging mechanisms to prevent artificial fragmentation during prolonged signal loss. Since optimal extraction parameters vary significantly across datasets, we further introduce a fully unsupervised, consensus-based parameter estimation framework. This framework derives baseline statistical scales directly from the raw trajectory data and identifies the optimal parameter configuration by maximizing structural agreement, thereby eliminating the reliance on annotated ground truth. We evaluate SSPE on the ExtraSensory dataset, a real-world mobility corpus with manually verified ground-truth staypoints, against the sliding-window implementation of Trackintel, one of the most widely adopted extraction baselines in the field. Empirical results demonstrate that SSPE achieves superior accuracy, yielding higher F1-scores while simultaneously improving computational efficiency over this standard.

0.19

Bernoulli On-Policy Distillation: Don't Ignore High-Confidence Student Tokens

cs.LG ·2026-11-01 ·Mingyu Cao, Yaokai Jin, Shiwei Liu, Lu Yin
code/data
Abstract · 14 pages, 7 figures, Code: https://github.com/duterscmy/Bernoulli-OPD

On-policy distillation (OPD) trains a student language model on its own trajectories using token-level supervision from a stronger teacher. We show that standard sampled reverse-KL OPD systematically suppresses updates to high-confidence tokens, such that their weak gradient contribution does not necessarily imply weak or uninformative teacher supervision. We introduce Bernoulli On-Policy Distillation (Bernoulli OPD), which directly matches the teacher and student probabilities for that token. This yields a bounded probability space correction that preserves teacher-student disagreement while avoiding the additional confidence-dependent suppression induced by vanilla OPD. Fixed-rollout analyses show that Bernoulli OPD substantially restores optimization signal to high-confidence tokens while remaining sensitive to teacher-student disagreement. We further analyze the resulting model behavior through confidence distributions, answer calibration, policy movement, and training dynamics, showing that Bernoulli OPD's gains cannot be explained simply by longer reasoning, larger optimization steps, or more uniform imitation of the teacher. Code is available at https://github.com/duterscmy/Bernoulli-OPD.

0.19

Partial Invariance for Classification under Structural Causal Models

cs.LG stat.ME ·2026-11-01 ·Austin Goddard, Yu Xiang
Abstract

Making predictions in an unseen environment from data collected across multiple training environments is challenging. This challenge is particularly pronounced when the target-generating mechanism varies across environments. In this work, we introduce environment-label separation, a form of partial environmental invariance that decomposes the target conditional distribution into an environment-invariant component and a label-invariant component. We show that this partial invariance enables estimation of the target conditional distribution in an unseen environment when using labeled training data together with unlabeled test features. Within the structural causal model framework, we provide sufficient conditions for environment-label separation. Building on these conditions, we develop an algorithm, dIMP-Classifier, which jointly minimizes an estimate of the test risk and encourages invariance across environments. We further provide finite-sample guarantees that characterize the test risk in terms of deviations from invariance, estimation error, and sample complexity. Experiments on simulated data and the real-world light tunnel Causal Chamber data demonstrate that the proposed approach can perform under interventions on the target mechanism and consistently outperforms the considered standard and invariance-based baselines.

0.19

TSSDA: An Autonomous Agent for Constructing a Time-Series Text-to-SQL Benchmark

cs.LG cs.DB ·2026-11-01 ·Zhihan Zheng, Shaoxu Song, Haitao Yuan, Minxiao Chen
benchmark
Abstract

As agents increasingly undertake data querying, Text-to-SQL serves as a foundational capability for database access. However, for text-to-SQL in time-series databases, high-quality benchmarks and large-scale training corpora remain scarce, limiting the systematic evaluation and domain adaptation of existing models and agents. Existing automated SQL generation approaches lack sufficient autonomy, leaving key challenges unresolved. First, numerous combinations of temporal syntax and functions complicate comprehensive coverage of valid queries. Second, as the corpus grows, generating valuable and diverse new data becomes increasingly challenging. Third, schema and data heterogeneity hinder reuse of temporal query structures, limiting generation efficiency. To address these challenges, we propose TSSDA, a Time-Series SQL (TSSQL) Generation Data Agent framework, and construct TERN, a comprehensive Time-sERies text-to-SQL beNchmark. First, to achieve broad coverage of valid queries, we systematically model TSSQL syntax and semantics to construct a data generation space. Next, to explore the space and improve diversity, we develop a Data Agent that autonomously orchestrates data generation. The agent comprises a planning and decision-making module and a query generation module, with online adaptation for continuous self-improvement. Finally, we introduce a temporal-aware memory that enables adaptive reuse of experience across heterogeneous data instances, improving generation efficiency. Comprehensive evaluations on TERN reveal the limitations of existing methods in performing Text-to-TSSQL tasks. Furthermore, extensive experiments demonstrate that TSSDA substantially improves coverage, diversity, and complexity.

0.18

SparseLift: Automatically Optimizing Opaque Sparse Paths in Industrial-Scale Recommendation Model Training

cs.LG ·2026-11-01 ·Changqing Shi, Anan Liu, Genbao Chen, Junchao Yang
benchmark
Abstract

The sparse path, comprising sparse operations and collective communication, is a critical component of large-scale recommendation models. However, these operations are often represented as opaque nodes in the training computation graph, hindering whole-graph optimization. We present SparseLift, a low-overhead compiler for the sparse path of large-scale recommendation training. It exposes sparse computation and communication to whole-graph analysis to expand the optimization space and improve training efficiency. By analyzing model structure and runtime information, SparseLift reconstructs opaque nodes as analyzable subgraphs, recovering the resource semantics of sparse computation and communication within the graph. Building on the recovered semantics, we design incremental cost models and a constraint-aware planner to automatically select and organize subgraph optimizations into an execution plan that is reused during subsequent steady-state training. We evaluate SparseLift on an industrial video-platform recommendation task with 100B sparse parameters and the public DLRM benchmark. SparseLift improves steady-state throughput for ordinary industrial training steps by 16.5\% over the native baseline. On DLRM, it achieves $1.60\times$ and $1.26\times$ model-execution speedups over DLRM-Baseline and TorchRec, respectively.

0.18

Learning Gradients of Homogeneous Functions

cs.LG eess.SPmath.OC ·2026-11-01 ·Mayur S. Gowda, Srinivasa Pranav, Shreyas Chaudhari, José M. F. Moura
theory
Abstract

Homogeneous gradient functions have widespread significance, including applications in maximum energy subspace selection and retrieval problems. Euler's homogeneous function theorem specifies how the potential associated with a positively k-homogeneous gradient field can be recovered directly from the field itself. We use Euler's theorem to build on GradNets and propose a homogeneous gradient network architecture that is guaranteed to correspond to the gradient of a homogeneous function. Then, we describe how monotone homogeneous GradNets universally approximate the gradients of k-homogeneous convex functions. The inductive bias of homogeneous GradNets enables them to outperform baseline ICNNs by up to 7dB in our max energy subspace projection experiment and by up to 13% higher accuracy, 98% fewer parameters, and 70% fewer FLOPS in our amortized inner product search experiment.

0.18

Direct Knowledge Transfer for Offline Q-Learning

cs.LG stat.ML ·2026-11-01 ·Zhanye Luo, Guangyan Gan, Lin William Cong
Abstract

Data-driven sequential decision making in marketing and healthcare can benefit from observations collected in related environments, particularly when data in the environment of interest are limited. Transferring such information requires accounting for differences in rewards and transition dynamics across environments. We study knowledge transfer for offline reinforcement learning through a joint estimation framework that directly integrates source information into target value-function learning. We propose Direct Transfer Fitted Q-Iteration (DFQI), which updates the target Q-function while simultaneously borrowing the source information through an L1 penalty that links the target coefficients to a weighted aggregate of source coefficients, with both determined jointly within each Bellman update. Using sieve approximation, we establish statistical guarantees for target estimation and policy learning in stationary discounted environments with common or heterogeneous transition kernels. Empirical studies further demonstrate the effectiveness of the proposed DFQI framework in transferring source information to improve policy learning with limited target data.

0.18

Beyond Imputation: Mask-Adaptive Conformal Prediction via Tree Embeddings on General Missing Data Mechanisms

cs.LG stat.ML ·2026-11-01 ·Jiarong Fan, Juhyun Park, Thi Phuong Thuy Vo, Nicolas J-B. Brunel
Abstract

Conformal prediction (CP) provides a distribution-free framework for uncertainty quantification with finite-sample marginal coverage guarantees. Yet, missing values introduce distributional heterogeneity, where marginal coverage can hide systematic undercoverage for certain missingness patterns. At the same time, existing methods based on distances or smoothness require a continuous representation of partially observed covariates. Although imputation restores continuity, it obscures the missingness patterns and introduces task-dependent overhead. We develop a new CP framework that is \emph{adaptive to missingness patterns}, without imputing covariates. We rely on data augmentation by feature embedding that accounts for missingness. Our framework represents each input with missing values by a tree embedding together with its raw missingness mask, and then learns a regularized quantile score function. Varying the choices for the penalty function and the augmentation enables us to characterize three practically relevant validity levels: exact coordinate-wise validity, exact validity on observed masks, and approximate missingness-conditional guarantees for general mechanisms (MCAR, MAR, MNAR). Experiments on synthetic and real datasets support these guarantees and show that the proposed method reduces miscoverage risk while maintaining informative intervals, especially in non-MCAR settings. These results highlight the practical value of uncertainty quantification that adapts to missingness.

0.18

Principled Human-AI Collaboration in Agentic Workflows

cs.AI stat.MLcs.LGstat.ME ·2026-11-01 ·Young Hyun Cho, Will Wei Sun
theory
Abstract · 44 pages, 4 figures. Includes supplementary material with proofs and experimental details

Organizations adopting AI face the challenge of deciding which outputs can be released without human review while meeting the quality standards of human experts. We develop a statistical framework for delegating tasks between experts and black-box generate-verify workflows, in which AI-generated answers are verified and revised over multiple rounds. Before deployment, human experts evaluate selected outputs, and their judgments are used to calibrate the verifier threshold and number of attempts. Our method selects the policy with the highest estimated automation rate among statistically approved candidates, without modifying the generator or verifier. We establish finite-sample control of the error rate among released outputs, characterize the expert evaluations sufficient to attain a target automation rate, and derive a minimax lower bound establishing rate optimality in the quality margin and required confidence. In a MultiHiertt financial question-answering case study, average held-out automation increases from 2.50% to 24.17% as expert evaluations accumulate, while released-output error rates remain around 1.20%. In contrast, heuristics that release answers after consistently high verifier scores either violate the 5% error tolerance or automate at most 1.30% of tasks. Synthetic experiments further examine these theoretical predictions under controlled settings.

0.18

Search-Aware On-Policy Self-Distillation with Sibling Alignment for Generative Recommendation

cs.LG cs.AI ·2026-11-01 ·Rongfeng Guo, Zhuang Zhuang, Ji Dai, Peng Zhao, Shijie Li +4 more
Abstract

Generative recommendation retrieves items by autoregressively generating semantic IDs, typically via beam search. On-policy self-distillation (OPSD) enhances training supervision by transferring privileged self-teacher knowledge at student-generated states. However, beam-search retention depends on competition among prefixes: visited states differ in importance, and initially reliable decisions may approach the pruning boundary as training evolves. Moreover, conventional distribution matching couples target-token confidence with preferences among competing continuations, potentially weakening relational supervision when the self-teacher strongly favors the target. To address these challenges, we propose Search-Aware On-Policy Self-Distillation (SA-OPSD), a framework that transfers privileged self-teacher knowledge to the states visited during the student's own beam search. Specifically, we introduce Evolving-Boundary Sampling, which prioritizes search-critical states by considering both current pruning-boundary proximity and initial-reference retention-margin decline under a controlled distillation budget. Furthermore, we develop Confidence-Decoupled Branch Distillation, which separates target-probability calibration from informative sibling relations to better preserve the teacher's preferences among competing branches. Experiments on Industrial, Office, and Instruments demonstrate strong performance for SA-OPSD.

0.17

CONFIRM: Governing Scientific Reporting for Trustworthy Agentic Neuroimaging

cs.AI cs.MA ·2026-11-01 ·Yuxiang Wei, Lijing Zhu, Qizhen Lan, Yanteng Zhang, Vince D. Calhoun
Abstract

Large language model (LLM) agents can autonomously run neuroimaging analyses from research questions to reported findings. Existing agentic systems use tools to check that code runs and reported values match its output, but rely on LLM judgment to decide which results warrant scientific reporting. We show that LLM judges endorse site-confounded findings even when the confound is presented as evidence. We present \textbf{CONFIRM}, a framework that separates running an analysis from deciding which findings to report. CONFIRM encodes each hypothesis as a machine-readable, pre-specified analysis plan that fixes the claim and its pass criteria before computing results. It then filters results through five deterministic gates (search history and multiplicity, confounds, power, analytic stability, and replication) under three configurable reporting standards, from least to most demanding: \textit{Screen}, \textit{Replicate}, and \textit{Full}. Failed claims receive diagnoses that guide follow-up analyses. To evaluate reporting decisions at the claim level, we introduce \textbf{NeuroClaimBench}, comprising 489 executable neuroimaging claims with frozen data, including literature-supported findings and 150 controls designed to violate individual methodological requirements. On NeuroClaimBench, CONFIRM falsely supports at most 3\% of controls, compared with 23--38\% for the baselines, while recovering up to 61\% of literature-supported claims. When auditing an existing agent, CONFIRM reduces false support on controls from 23\% to 0\% while retaining the agent's literature-supported claims. Its diagnostic feedback also produces more supported revisions than Self-Refine with fewer LLM calls. Together, CONFIRM and NeuroClaimBench show that basing reporting decisions on pre-specified analysis plans and deterministic gates is essential for trustworthy autonomous scientific discovery.

0.17

I-PTC: Interactive Programmatic Tool Calling for Stateful Tool-Augmented Agents

cs.AI cs.LG ·2026-11-01 ·Huanzhi Mao, Chengkun Cao, Shuo Yuan, Joseph E. Gonzalez
benchmark
Abstract

Large language models are increasingly capable of writing and executing code, suggesting a natural paradigm for tool-augmented agents: expressing tool use as programs rather than isolated calls. Programmatic tool calling (PTC) captures this idea through loops, batching, abstraction, and structured computation, but existing PTC approaches are largely stateless, limiting their ability to reuse intermediate results and adapt over time. This is problematic for real-world tasks, where tool environments and problem-solving processes are inherently stateful. We therefore present I-PTC, an interactive framework that executes stateful PTC: models issue code snippets that call tools, maintain reusable state, and evolve the tool interface through API corrections. To evaluate the effectiveness of I-PTC and the general PTC setting, we introduce PTC-Bench, a benchmark of long-tail, production-style user requests whose difficulty comes from large initial environments, many dependent API interactions, extended planning, and recovery across follow-ups and transient failures. Experiments on PTC-Bench show that I-PTC improves performance on complex multi-step tasks with large tool sets while substantially reducing token consumption over conventional tool-calling baselines and the original PTC baseline, highlighting stateful programmatic tool calling as a promising direction for scalable tool-augmented agents.

0.17

OC-OPD: Observation-coordinated Supervision for On-Policy Distillation

cs.LG cs.AImath.OC ·2026-11-01 ·Ruoran Xu, Yilun Zhao, Ziyu Chen, Arman Cohan
code/data
Abstract

Reverse and forward Kullback--Leibler (KL) divergences provide complementary supervision in on-policy distillation: reverse KL refines the student's active token rankings, and forward KL allocates mass to teacher-supported candidates. At student-generated prefixes, we form a four-observation matrix by treating the student and teacher as both candidate sources and evaluators. The diagonal self-observations measure uncertainty, while source-centered cross-observations recover the two KL objectives. We establish an exact identity equating the Fisher inverse-metric coupling of their logit gradients with the sum of these contrasts, yielding a common local descent direction for distinct positive distributions. Observation-coordinated on-policy distillation (OC-OPD) uses observations to set preferences, closed-form bargaining to coordinate gradients, and distributional discrepancy to scale the update. Across three heterogeneous student settings (non-thinking generation, thinking generation, and base-model initialization), OC-OPD achieves strong gains in mathematical reasoning and transfers effectively to out-of-distribution scientific question answering and code generation. Our code is publicly available at: https://github.com/RuoranXu/OC-OPD.

0.17

Physics-Informed One-Step Generation on Long-short flow maps

cs.LG ·2026-11-01 ·Zhiqi Li, Duowen Chen, Jinjin He, Bo Zhu
Abstract

Physics-informed generative models should produce samples that are both data-aligned and physically valid, yet existing methods are often iterative and too slow for practical use. Achieving this with a one-step flow-map model is difficult because both data-distribution and physics-constraint learning are enforced only indirectly through surrogate objectives, making their joint learning fragile for long-range flow maps. We propose \textbf{Physics-Informed Long-Short Flow Maps with Direct Guidance (LSFM-DG)}, which makes direct target computation feasible by decomposing one-step generation into a long flow map for global evolution and a near-terminal short flow map with computable closed-form solutions and physics potentials. As the short flow map shrinks to zero, it provides both direct distributional and physical targets for learning the long flow map. This yields a unified one-step flow-map model that removes Jensen's gap, improves training stability, and achieves stronger data fidelity and physical consistency without distillation.

0.16

Direct Covariance-to-Beamformer Mapping for Low-Latency Hybrid ISAC Beamforming with Subsequent DoA Estimation

cs.LG eess.SY ·2026-11-01 ·Anjali K. S., Manoj Kumar Panda, Chandramani Singh
Abstract

Integrated sensing and communication (ISAC) systems typically employ a two-stage beamforming pipeline, in which target directions are first estimated from the received covariance matrix and then used to compute the beamforming weights by solving non-convex optimization problems. Although effective, this approach introduces processing latency that can limit beam adaptation in dynamic environments. This letter proposes a beamforming deep neural network (BFDNN) for direct covariance-to-beamformer learning in ISAC systems. BFDNN is a compact regression network that maps covariance-matrix features directly to the hybrid beamforming weights produced by a hybrid ISAC beamformer, eliminating the direction-of-arrival (DoA) estimation stage during inference. Compared with MUSIC, classification-DNN, and regression-DNN based beamforming pipelines, BFDNN generates comparable beam patterns and higher effective spectral efficiencies while reducing the beamforming prediction latency by 25-30 times on the evaluated platform, demonstrating the potential of direct covariance-to-beamformer learning for low-latency ISAC beam updates. Moreover, the target directions can be inferred indirectly from the generated radiation pattern, yielding DoA accuracy competitive with that of dedicated learning-based methods in the evaluated scenario.

0.16

Robust EEG-based Emotion Recognition from Coarse Trial-level Labels under Intra-trial Affective Variability

cs.LG cs.AI ·2026-11-01 ·Xiaocong Zeng, Buxuan Wang, Xin Yang, Pei Huang, Dongyang Kuang
Abstract

EEG-based emotion recognition has become an active research area in brain-computer interfaces. However, fine-grained affective annotation is costly, so most datasets provide only a global label for each long EEG trial. For practical emotion recognition, trials are commonly divided into short temporal windows, with the global label propagated to all windows for training. This practice treats trial-level emotion as temporally uniform, although affective responses may fluctuate within a trial. Consequently, windows that weakly reflect or deviate from the global emotional state may receive unreliable labels, posing a key challenge to robust EEG-based emotion recognition. To address this challenge, we propose VARY, a variability-aware robust learning framework for EEG-based emotion recognition from coarse trial-level labels. VARY incorporates a Two-stage Sample Selection (TSS) strategy to identify EEG windows with reliable emotional labels, and introduces MixAug, a data augmentation method tailored for EEG to improve model robustness under limited and unreliable supervision. We comprehensively evaluate VARY under the setting of label-noise learning on two public datasets, using three network architectures and multiple label-noise settings. Our method demonstrates superior performance, achieving a maximum accuracy gain of 11.54% on the DREAMER dataset and 5.39% on the SEED dataset compared with the best-performing baseline, highlighting the effectiveness of VARY in learning from unreliable trial-level labels under intra-trial affective variability.

0.16

Watch Your Step: Adam's Margin Maximization Bias Depends on the Learning Rate Schedule

cs.LG stat.ML ·2026-11-01 ·Eitan Gronich, Gal Vardi
theory
Abstract

We study the implicit bias of Adam and its similarity to sign gradient descent on homogeneous models in binary classification tasks. Perhaps surprisingly, we uncover that these depend on the learning rate schedule. While previous works studied strongly decaying learning rate schedules $η(t)=\mathrm o \left(t^{1/L-1}\right)$, where $L$ is the homogeneity degree, and showed $\ell_\infty$-margin-maximization biases, we focus on the more common complementary regime, including constant and weakly decaying schedules. For a polynomial schedule $t^α$ and homogeneity degree $L$, we show that when $α> \frac 1L - \frac12$, Adam has no trajectory-independent margin-maximization bias for deep linear networks (in the KKT sense): first, there is no non-trivial, locally Lipschitz, convex implicit regularizer in this setting, including any squared norm. Second, on width-$1$ networks, the result holds without requiring convexity, implying that any implicit regularizer for wide networks must be trivial when restricted to a width-$1$ network. Nevertheless, we positively characterize the implicit bias in this regime as maximizing a trajectory-dependent weighted-$\ell_2$ margin for any $L$-homogeneous model. The fundamental mechanism behind our results is convergence of Adam's preconditioner to a trajectory-dependent vector, which also allows showing that the limiting parameter direction depends on initialization. In the intermediate regime $\frac 1L - 1 \leq α\leq \frac1L - \frac12$ we prove an $\ell_\infty$ bias, conditional on gradient directional convergence to a full-support vector. Of independent interest, we establish a loss decay rate for full batch Adam on general homogeneous models, whose leading order depends only on the second-moment parameter.

0.15

Improved Sample Complexity Analysis For First Order Bilevel Reinforcement Learning

cs.LG cs.AI ·2026-11-01 ·Mudit Gaur, Naman Saxena, Vaneet Aggarwal
theory
Abstract

Bilevel reinforcement learning optimizes an upper-level objective through the policy response to a lower-level reinforcement learning problem. Such problems arise when learning a reward model while adapting a policy to that reward. However, a gap remains between sample-complexity guarantees and practical actor--critic training. Tabular analyses restrict applicability to finite spaces, while continuous-state analyses often impose a uniform Polyak--Łojasiewicz (PL) condition. We show that this condition need not hold even for smooth positive policies with exactly representable values. We develop a single-loop, fully first-order method for continuous states and finite actions with specified multilayer neural actors and a persistently trained nonlinear critic built from fixed features. Under explicit initialization, regularity, feature-excitation, and value-realizability conditions, fixed reference-KL regularization yields local curvature, and we prove containment of the coupled iterates. Shared trajectories control sampling noise, while paired critic analysis controls normalized critic-error differences as the response perturbation shrinks. A joint descent and tracking argument establishes an improved $\widetilde{\mathcal O}(\varepsilon^{-2})$ sample complexity for $O(\varepsilon)$ expected squared gradient norm of the upper objective induced by the selected regularized policy response, including critic training.

0.15

When Is Personalized Federated Learning Worth Its Deployment Cost?

cs.LG ·2026-11-01 ·Beyazit Bestami Yuksel, Ayse Yilmazer Metin, Emrah Dikbiyik
published
Abstract · 6 pages, 2 figures. Accepted and presented at UBMK 2026

Personalized federated learning (PFL) may improve client-level performance, but deeper personalization increases deployment cost. We formulate personalization-depth selection as a utility-based empirical decision problem in which client utility equals Macro-F1 minus a cost-sensitivity weight times deployment cost, over six personalization depths labeled P0 through P5. We compare fixed policies with an offline per-client Oracle reference and report allocation maps over Dirichlet data heterogeneity and cost sensitivity. Although evaluated on this hierarchy, the formulation can accommodate any finite candidate set. On CIFAR-10, CIFAR-100, and a PTB-XL subset, personalization shows empirical, setting-dependent decision regimes rather than a universal benefit: higher cost sensitivity shifts Oracle mass toward the global-only depth P0; low cost sensitivity can favor deeper actions. Patterns are grid observations (three seeds per cell), not theoretical boundaries. Best-uniform is competitive with Oracle assignment (6.5 percent plus or minus 7.8 percent relative regret; 95 percent confidence interval 5.9 to 7.1), whereas random and personalize-all incur much larger regret (189.3 percent plus or minus 367.1 percent and 409.0 percent plus or minus 803.2 percent). Under the same utility, Ditto full-model personalization is substantially suboptimal once cost is counted. The contribution is an offline characterization of when personalization depth is worth its cost, not a new PFL optimizer or deployment-time scheduler.

0.15

HQNN-CL: Regime-Aware Hybrid Quantum Neural Network with Continual Learning for Financial Forecasting

cs.LG cs.AI ·2026-11-01 ·Muhammad Atif Saeed, Akhtar Jamil, Subhan Ullah
Abstract · 20 Pages, 8 Figures

Financial time-series forecasting is still difficult owing to the non-stationary nature of the market, regime changes, and catastrophic forgetting, which makes conventional deep learning models hard to adapt in the long term. To achieve uncertainty-aware robust financial forecasting in the face of changing market conditions, this study proposes a Hybrid Quantum-Classical Neural Network with Continual Learning (HQNN-CL). The proposed framework combines Gaussian Hidden Markov Model (HMM) based regime detection, BiLSTM temporal encoding, adaptive-depth parameterized quantum circuits, Elastic Weight Consolidation (EWC), experience replay, and probabilistic forecasting in a single architecture. The framework was tested on four real-time stock datasets: Microsoft, Systems Limited, Teradata and Tata Consultancy Services, and compared with seven state-of-the-art forecasting models, including recurrent and Transformer-based models. The experimental results show that HQNN-CL has outperformed other methods on the Systems Limited (RMSE = 3.0149) and Teradata (RMSE = 0.8909) datasets and is within 0.32% and 0.40% of the best baseline method for Microsoft and Tata Consultancy Services, respectively. The proposed framework also generates well-calibrated probabilistic forecasts within the range of 89.5% to 96.5% empirical coverage rate, has zero catastrophic forgetting (BWT = 0.0000) for all sequential learning tasks, and consistently shows positive forward transfer. Ablation and statistical significance analyses also validate the performance of adaptive quantum feature learning and continual adaptation under heterogeneous market conditions, regime-wise.

0.14

Meta-memorization and memorization scaling laws in transformers

cs.LG cs.AI ·2026-11-01 ·Alex Nguyen, Kenneth A. Norman, Gautam Reddy
Abstract

Larger transformers can memorize more data, with a capacity (in bits) that scales roughly linearly with the number of parameters. Counterintuitively, larger models also memorize faster, requiring fewer repetitions of a sequence during training before it is memorized. We reproduce both memorization scaling laws in transformers trained on random binary sequences, showing that rapid memorization can arise even when data contain no semantic or compositional structure. Rapid memorization is due to a meta-memorization effect, where models trained to memorize a large set of random sequences learn to memorize novel sequences more quickly. The rate $r(t)$ at which models memorize grows with training time $t$ as a power-law $r(t) \sim t^β$ with a fixed exponent $β$ but an increasing upper cutoff across model scales. Meta-memorization enables even one-layer transformers to few-shot memorize a 100-bit sequence. This effect is partly explained by the formation of structured attention heads that convey position-invariant sequence information to MLPs where memorized sequences are stored. Meta-memorization is recapitulated in a simplified model of a dense associative memory where retrieval is constrained by an attentional bottleneck. Our results suggest transformers, through pre-training, acquire reusable mechanisms that enable highly efficient storage and retrieval of memories.

0.14

A Self-Exploring and Self-Evolving Paradigm for Adaptive Edge AI Development

cs.AI cs.SE ·2026-11-01 ·Leming Shen, Qiang Yang, Xinyu Huang, Zijing Ma, Yuanqing Zheng +1 more
Abstract

Large Language Models (LLMs) exhibit powerful code generation abilities, fostering agentic systems to automate edge AI application development. However, existing agents typically operate via fixed agentic workflows and predefined design spaces, struggling to handle rapidly evolving demands of real-world scenarios (e.g., heterogeneous data modalities, diverse task pipelines). Unlike prior systems that treat LLMs mainly as code generators following rigid workflows, we fully exploit their reasoning capabilities to allow agents to determine their own workflows on the fly. To this end, we propose SE2, a self-exploring and self-evolving paradigm for adaptive edge AI development. Inspired by human learning processes, we distill development into four atomic operations (retrieve, reason, code, test), and empower agents to explore workflows by dynamically and adaptively selecting an appropriate next operation based on the current context. Extensive evaluations show that, with adaptive workflows, the synthesized programs achieve up to 26% higher accuracy and consume 28% lower resources than state-of-the-art baselines, while maintaining superior adaptability to unforeseen edge AI scenarios. This highlights the potential of reasoning-driven adaptive agentic workflow in edge AI development.

0.14

Learning AI Agents, Market Makers, and Stress in Automated Markets

cs.LG econ.EMecon.THq-fin.TR ·2026-11-01 ·Irene Aldridge
Abstract · 15 pages

Automated markets increasingly route order flow from learning trading agents through automated market makers (AMMs). We model trader archetypes as AI agents defined by their reward functions and analyze the feedback they create on a constant-product AMM. Herd-dominated and arbitrage-dominated markets of equal size have identical return distributions and turnover, so no rule based on volatility, value-at-risk, stress frequency, or volume can tell them apart; yet their stress episodes differ in length by a factor unbounded in the feedback-to-noise ratio $β/s$. Serial statistics identify the feedback sign from $Θ((s/β)^2\log(1/δ))$ observations, and this rate is tight. When agents trade together, profit-seeking agents ride herds rather than lean against them, leading to all-or-nothing participation. Spread-earning market makers restore stability but make liquidity fragile, with discontinuous collapse and hysteresis; maker--taker pricing, not a uniform fee, is the effective instrument.

0.14

Value Forcing: Aligning Return Distributions at All Noise Levels

cs.LG cs.AI ·2026-11-01 ·Chongyi Zheng, Zhiyuan Zhou, Qiyang Li, Benjamin Eysenbach, Sergey Levine
position
Abstract

Learning accurate value functions is central to reinforcement learning (RL), yet remains challenging because supervision is limited and indirect: ground-truth values are unavailable, and each transition in the data only provides an immediate reward that can be used to construct bootstrapped targets as supervision. This challenge is particularly pronounced when data is limited or reward signals are sparse, as is common in real-world applications such as robotics. We argue that the key to accurate value learning in this limited data regime is extracting richer learning signals from the available data. In this paper, we introduce Value Forcing, a new approach that uses flow matching to estimate return distributions and enables accurate value learning from limited data. Our key insight is that we can extract richer learning signals by enforcing Bellman consistency at every noise level. Empirically, Value Forcing improves the accuracy of estimating the return distribution, and prevents the distribution from collapsing to a single value, as occurs with prior methods. We instantiate Value Forcing with two practical value-learning algorithms: VF, an actor-critic method, and IVF, an implicit value-learning method. We evaluate both algorithms across $26$ long-horizon tasks in the offline RL setting and find that they outperform prior value-learning methods, with particularly large gains when data is limited. Our project website is available at https://chongyi-zheng.github.io/value-forcing.

0.13

Retrain, Alert, or Rollback? A Three-Response Decision Framework for Production ML Under Concept Drift

cs.LG cs.SE ·2026-11-01 ·Ali Sadeghian, Alireza Shafaei, Hamid Hadibeik, Babak Hossein Khalaj
Abstract · 51 pages, 3 figures. Submitted to Data Mining and Knowledge Discovery

When a deployed model degrades under concept drift, an operator can retrain it, raise an alert and monitor, or roll back. Yet most drift-adaptation methods consider only retraining. We present RAR (Retrain-Alert-Rollback), a state machine that gives each response an explicit trigger and cost. Retraining is gated on drift detection, label availability, and measured degradation. Every retrained model must pass a two-stage canary before promotion, and corruption-triggered and quality-triggered rollbacks enter different hold states. Under a prospectively specified protocol we compare RAR with six baselines, including a cost-aware retraining policy, on 13 streams: eight real public streams and five semi-synthetic streams derived from them. RAR does not dominate. At low retraining cost, retrain-heavy baselines lead, and no policy's rank differs significantly from RAR's. At higher cost, RAR beats detector-triggered retraining, periodic retraining, and no response simultaneously on four streams, each within a bounded cost window. Its advantage over detector-triggered retraining is consistent with a latency-feasibility condition stated in advance. Against the cost-aware policy, results are heterogeneous. RAR retrains before true drift onsets one eighth to one third as often as detector-triggered retraining, but at lower net utility on most semi-synthetic streams. On H2's designated corruption stream (a registered follow-up), rollbacks fire yet RAR loses to retraining baselines. At $ρ= 1$, removing the canary, and thus rollback, raises utility on 11 of 13 streams. On a post-hoc follow-up stream, cause-specific hold release resumes adaptation where a single cooldown does not. We provide the harness, configurations, and per-cell results for reproduction.

0.13

Beyond Binary Detection: Type-Aware Hallucination Analysis and Hierarchical Detection in Financial Question Answering

cs.AI q-fin.CP ·2026-11-01 ·Xiongyu Chen, Sizhen Zhu, Zhuoer Wang
benchmark
Abstract

Large language models are increasingly used to answer questions over financial filings, where an unfaithful answer can mislead investment, compliance, or client decisions. Hallucination detectors are a common safeguard, but they are usually evaluated with a single binary score that reveals neither which kinds of errors analysesed nor why. This paper studies hallucination detection on PHANTOM, a long-context benchmark built from SEC filings, at the level of hallucination type. We first analyze 15,443 paired hallucinated and faithful answers without any detector and find that hallucinated answers are more than twice as long as faithful ones, and that lexical grounding separates the two classes only because faithful answers are bimodal. We then generate hallucinations in nine financially motivated categories, confirm each with a separate verification pass, and diagnose an LLM-based detector on 4,086 usable examples. Generation success varies from 66.3% to 100% across categories. The detector reaches 80.2% binary accuracy but only 44.7% subtype accuracy, never assigns two of the nine categories, and misses between 5.3% and 46.8% of hallucinations depending on category; these misses are unrelated to input length but rise with how far an answer departs from the correct one. Guided by this diagnosis, we propose a two-stage hierarchical detector that first decides whether an answer is hallucinated and then assigns a subtype, raising binary accuracy to 95.6% and subtype accuracy to 56.1%. The results show that type-level evaluation exposes detector weaknesses that aggregate metrics conceal.

0.13

Route-Guided CBF-QP Repair for Safe Vision-Language-Action Models

cs.AI cs.RO ·2026-11-01 ·Chan Park, Sunwoo Hong, Kirak Kim, Yoonjeong Park, Alexander W. Olson +1 more
code/data
Abstract

Vision-language-action (VLA) models can generate task-oriented robot actions, but they do not guarantee safe execution in obstacle-rich manipulation scenes. Although CBF-QP safety filters can locally project unsafe actions into a safe set, such correction may push the robot away from the learned task trajectory and leave the VLA policy without a reliable recovery direction. We propose Route-Guided CBF-QP Repair, a post-hoc repair module that uses the short-horizon action chunk predicted by the VLA to detect blocked nominal trajectories and generate entry-exit-rejoin guidance around obstacles. The selected route is added to the CBF-QP objective as a soft guidance term, while the CBF safety constraint remains hard. On SafeLIBERO, our method improves collision avoidance, task success, and SafeSuccess compared with AEGIS. Our project page is available at GitHub.

0.13

ReCoG: Relative Node Contexts for Multi-Domain Graph Pre-training

cs.LG ·2026-11-01 ·Yupeng Tong, Lianze Shan, Jitao Zhao, Dongxiao He
Abstract · 13 pages, 7 figures, 8 tables

The success of foundation models in language and vision has motivated Graph Foundation Models (GFMs), which learn transferable knowledge through multi-domain graph pre-training. Existing methods advance GFMs through feature unification and structural learning, and in doing so they rely on the raw graph structure to organize node contexts. However, we observe that removing a large proportion of edges from source graphs during pre-training improves cross-domain transfer accuracy across four representative methods. Our theoretical analysis identifies the raw graph structure as a source of conflict in cross-graph model sharing. We therefore advocate defining node contexts using criteria consistent across graphs. Achieving this requires (1) constructing node contexts consistently across graphs without a shared positional structure, such as the token order of text or the pixel grid of images, and (2) retaining the structural and neighborhood information of the original graphs. To this end, we introduce Relative Contexts for Graphs (ReCoG) for multi-domain graph pre-training. Specifically, we first construct attribute contexts to provide a comparison rule independent of the raw graph structure. Furthermore, we construct structural contexts based on common neighbors and semantic contexts based on neighborhood attribute summaries to retain the topological and neighborhood information of the original graphs. Since all graphs share these construction rules, we pre-train a single graph encoder by reconstructing node attributes over these contexts, without domain-specific modules. We conduct extensive experiments on 18 target datasets, which demonstrate the effectiveness of ReCoG. In particular, ReCoG exceeds the strongest baseline by 7.61 percentage points in mean 1-shot cross-domain node classification accuracy and by 4.47 points in cross-type transfer to heterophilous graphs.

0.13

To Align Long, Plan Short: Simulating Diverse Profile-Aligned Students by Realizing Short-Term Plans in Long-Horizon Learning Environments

cs.AI cs.CLcs.CY ·2026-11-01 ·Bahar Radmehr, Jiaxu Zhao, Fatma Betül Güreş, Adish Singla, Tanja Käser
Abstract · 33 pages, 3 figures, 20 tables

Simulated students offer a scalable way to test learning platforms and AI tutors before deployment with real students and to augment scarce data for studying learning at scale. Existing LLM-based simulators primarily target short, language-centered interactions. In open-ended learning environments, however, student behavior emerges from patterns across long-horizon interaction trajectories, where turn-level prompting can drift from an assigned behavior profile. We introduce ALPS (Align Long-term by Planning Short-term), a hierarchical simulated student conditioned on observable, environment-specific behavior profiles. Rather than selecting every action independently, ALPS alternates between planning profile-aligned task segments and generating the corresponding actions. We evaluate ALPS in both data-sufficient and data-scarce settings, where supervision comes either from real student trajectories or from a teacher LLM. Across both settings, we compare in-context learning (ICL), supervised fine-tuning (SFT), reward-ranked fine-tuning (RAFT), and their sequential combination (SFT followed by RAFT). Experiments in an inquiry-based physics game and a pharmacy simulation evaluate behavioral density and coverage against held-out real data, and profile consistency using a human-aligned LLM judge. Across both settings, ALPS consistently outperforms turn-level generation, achieving comparable density and judge pass rates under scarce supervision, although coverage remains lower. These results show that hierarchical planning enables faithful, profile-aligned simulation over long interaction horizons, even with limited real student data.

0.13

Near-Optimal $\sqrt{K}$-Regret for Online Learning in Uninformed Markov Games

cs.LG stat.ML ·2026-11-01 ·Junyan Liu, Haipeng Luo, Lillian J. Ratliff, Zihan Zhang
Abstract

We study online learning in two-player uninformed Markov games, where the opponent may follow an arbitrary, unknown policy in each episode and the learner does not observe the opponent's actions. [TWYS21] showed that achieving sublinear external regret requires an exponential dependence on the episode length $H$. Motivated by this limitation, they studied the weaker notion of Nash-value regret and obtained an $\widetilde{O}(K^{2/3})$ regret bound, where $K$ is the number of episodes. More recently, [LLZR26] introduced a stronger notion called empirical Nash-value regret and achieved an $\widetilde{O}(\sqrt{K})$ bound when the opponent's non-stationarity is moderate, including the case where the opponent follows a fixed policy. However, whether the $\sqrt{K}$ rate is achievable in the worst case remained open. In this paper, we resolve this open question by developing an algorithm that achieves $\widetilde{O}(H^{3/2}S\sqrt{AK})$ empirical Nash-value regret against arbitrary opponent policies, where $S$ is the number of states and $A$ is the number of learner's actions. Specifically, our algorithm combines global and local planning by maintaining a collection of candidate predictions of future losses, each associated with a policy. Global planning learns which candidates to favor across episodes, while local planning combines their policies differently at each state to make decisions. A key ingredient is to coordinate these two components to control the discrepancy between local loss estimates and the total loss of the executed policy, using only the learner's observed trajectories.

0.12

TriPhysBench: Evaluating Physical Intelligence for Embodied Agents

cs.AI cs.CV ·2026-11-01 ·Zhongxiang Sun, Chenghao Yang, Haojie Ding, Boheng Zhang, Fan Yang +2 more
benchmark
Abstract

Embodied agents need physical intelligence to understand their surroundings through physical perception, physical knowledge retrieval, and physical reasoning. These capabilities must work together to interpret events and make quantitative judgments. Existing benchmarks advance physical understanding and general-purpose tool use, yet the coordination of these capabilities in physical scenarios remains insufficiently characterized. Inspired by Popper's Three Worlds, we introduce TriPhysBench, a benchmark of 410 questions from 200 videos that brings physical perception, physical knowledge retrieval, and physical reasoning into a shared evaluation setting. Its measurement-grounded and knowledge-grounded subtasks impose complementary demands on obtaining observations and acquiring external specifications, then applying them to quantitative problems. Among six representative vision-language models, the best achieves an answer accuracy of only 50.38%, revealing limited performance on tasks that require these capabilities to work together. We further introduce TriPhys, a framework that learns a unified policy for physical perception, knowledge retrieval, and quantitative reasoning from complete tool-integrated trajectories. The policy uses intermediate reasoning to guide retrieval and computation, then incorporates tool feedback into subsequent physical inference. On 110 questions from videos excluded from training, TriPhys improves answer and solution scores across three Qwen configurations, with a 17.7% relative gain in answer accuracy for TriPhys-27B over its Qwen3.6-27B backbone. This work provides a benchmark and a learning pathway toward embodied agents that move beyond recognizing events to understanding and reasoning about the physical processes behind them.

0.12

MAC-Bench: Benchmarking Multi-Agent Collaboration via Interconnected Tasks

cs.MA cs.AIcs.CL ·2026-11-01 ·Zerui Chen, Qinggang Zhang, Qichuan Liu, Zhishang Xiang, Zhimin Wei +4 more
benchmark
Abstract

Large language model (LLM)-based multi-agent systems (MAS) tackle challenging tasks by dividing work among agents and coordinating their contributions toward shared objectives. Evaluating these systems therefore requires benchmarks that assess not only task-solving performance but also the ability to handle interdependent work. However, existing benchmarks often emphasize task difficulty without substantial collaboration requirements or evaluate only a limited range of predefined collaboration patterns. In this paper, we introduce MAC-Bench, a comprehensive benchmark for evaluating multi-agent collaboration through interconnected tasks. Rather than imposing predefined agent roles or communication protocols, MAC-Bench embeds collaboration requirements directly into task dependencies and shared-resource constraints. Specifically, we mine reusable collaboration structures from selected benchmark tasks at two levels. Local collaboration patterns capture recurring interactions among subtasks, while global task structures characterize how these interactions contribute to complete workflows. Through coarse-to-fine composition, these structures are recombined into task graphs combining different collaboration requirements. Tasks are instantiated in executable environments with automated verification of partial progress and complete task resolution.

0.12

Rare Event Constrained Reinforcement Learning with Large Deviations

cs.LG stat.MLeess.SYmath.OC+2 ·2026-11-01 ·Rahul Misra
theory
Abstract

We study finite-horizon reinforcement learning with a safety constraint on the probability of reaching a failure set. In contrast to the well-studied safe reinforcement learning literature, we consider problems where the probability of reaching the failure set is vanishingly small, but the consequences of reaching the failure set are catastrophic. Direct simulation and estimation of the optimal policy is naturally ineffective when the failure probability is extremely small. Moreover, on a real system the controller can only choose actions, and this is not enough to make failures frequent. In contrast to the rare event simulation literature, our goal is not to estimate the probability of the rare event occurrence, but to upper bound it such that the failure constraint is satisfied. Thus, instead of sampling failures, we use the large deviations principle. We define a damage variable that on exceeding a certain threshold indicates failure. We study the cumulant generating function (CGF) as its positive root provides an upper bound on the failure probability via Cramer's theorem. We extend this approach to Markov decision processes with hitting times through state-action CGFs, and give finite-horizon and state-dependent versions of the bound. The state-action CGFs do not depend on the policy and can be estimated from regular transitions, so the bound can be checked for any policy a priori. This gives a policy improvement algorithm in which every policy satisfies the failure constraint with high probability.

0.12

Finite-Time Online Sample Complexity for Conditional Diffusion Models

cs.LG cs.AI ·2026-11-01 ·Mudit Gaur, Prashant Trivedi, Sasidhar Kunapuli, Amrit Singh Bedi, Vaneet Aggarwal
Abstract

We study the finite-time training and sampling error of conditional diffusion models trained by online stochastic-gradient denoising score matching. Rather than assuming access to an exact empirical or population risk minimizer, we analyze the score network produced after a finite number of stochastic-gradient updates under an explicit localized Polyak--Łojasiewicz (PL) condition, stochastic-gradient regularity, and a trajectory-containment assumption. Our analysis separates the network approximation complexity from the online training-sample budget and propagates both errors through the reverse diffusion process. Under anisotropic smoothness, a joint intrinsic-dimensional structure for the data and conditioning variables, sub-exponential tails, and explicit approximation-class compatibility, we obtain a conditional score approximation error of order \[ \widetilde{\mathcal O}\!\left(M^{-η_{\mathrm{app}}}\right), \qquad η_{\mathrm{app}} = \frac{2α_x} {2α_x+d_x'+(α_x/α_y)d_y'}, \] where \(M\) denotes the approximation complexity, \(d_x'\) and \(d_y'\) are the intrinsic response and conditioning dimensions, and \(α_x,α_y\) are their corresponding smoothness exponents. For \(N=RB\) online denoising samples used by SGD, the finite-time optimization error is controlled explicitly in terms of the PL, smoothness, gradient-noise, and initialization constants. Combining these ingredients with reverse-process stability yields a total-variation bound that decomposes into terminal-initialization, time-discretization, approximation/localization, and finite-time optimization errors.

0.11

Spectral Signatures of Memorization Basins in Diffusion Trajectories

cs.LG ·2026-11-01 ·Achref Jaziri, Martin Mundt, Ahmed Imtiaz Humayun, Visvanthan Ramesh
Abstract · 44 pages, 15 Figures

Diffusion models are known to reproduce individual examples from their training data as a form of memorization with implications for generalization, privacy, and downstream use. Recent work demonstrated that memorized training samples reside inside contiguous regions of the data/latent space known as basins of memorization -- once a denoising trajectory enters a basin, under finite perturbations it irreversibly reproduces a memorized example. Prior empirical work on such basins in trained models has primarily focused on classifier-free guidance based detection and mitigation strategies in text-to-image models, leaving basin entry and their formation largely uncharacterized. In this paper, we present a geometric characterization of memorization basins for conditional, unconditional, and rectified-flow-based diffusion models. We observe a characteristic signature in the local spectrum of the function geometry that distinguishes basin entry: the emergence of a dominant singular mode followed by a sharp spectral recovery. This signature appears at two timescales: along denoising trajectories during inference and over the course of training. We further investigate whether these geometric signatures can be used to intervene on memorization. During training, we suppress memorization by acting on the dominant mode. At inference, we develop methods to identify trajectories entering memorization basins and steer them away. We find that these interventions remain effective even with low-dimensional approximations of the underlying geometry, and demonstrate that geometry-informed steering extends to large-scale text-to-image diffusion models. Together, our results corroborate that local spectral geometry provides a common basis for characterizing memorization and for suppressing, detecting, and steering away from it across training and generation.

0.11

Delivered, Not Used: Evaluating Agent Memory at Its Point of Use

cs.AI ·2026-11-01 ·Changran Xu, Zeju Li, Qiang Xu
Abstract

Task success, retrieval quality, and cost do not show whether an LLM agent used a delivered memory or what followed from that use. We introduce MemUse, which evaluates memory at its point of use in recorded trajectories: each memory delivery is compared with what the agent could already see and followed through the agent's later actions and their consequences. An agentic judge rates each delivered entry's relevance, accuracy, novelty, influence, and observed effect; programmatic checks enforce complete reading of cited evidence, and fixed rules aggregate the ratings into verdicts linked to memory histories and delivery cost. On 8,000 SWE-bench Verified episodes from three memory systems, resolve rates stay within 0.4 percentage points of a no-memory baseline, yet MemUse shows that most delivered entries were irrelevant, most relevant entries went unused, and in two systems most deliveries restated content the agent could still see. It also surfaces effects that outcomes hide: 23.2\% of episodes with beneficial memory use still failed, and 5 of 16 with harmful use still succeeded.

0.11

EMBERS: Empirically Motivated Robustness Testing of Deep Research Systems via Defect-Guided Fuzzing

cs.AI cs.LG ·2026-11-01 ·Shuochuan Li, Zan Wang, Dong Wang, Yuhang Jiang, Huajin Cui +3 more
Abstract

Deep Research (DR) systems extend large language models into autonomous agents that decompose complex information needs, iteratively search the web, and synthesize evidence to answer challenging questions. However, their multi-stage execution may introduce robustness risks, as variations in user questions or intermediate search inputs may propagate across research stages and lead to incorrect answers. Existing evaluations focus on task-solving capabilities or robustness against external evidence disturbances, leaving intrinsic weaknesses in DR execution insufficiently explored. Thus, we conduct an empirical study of 1,932 execution trajectories from three representative DR systems and identify nine recurring defect categories across subquestion formulation, online search, and final reasoning. Guided by these findings, we propose EMBERS, an empirically motivated, defect-guided fuzzing framework for robustness testing of DR systems. Starting from correctly answered seed questions, EMBERS generates stage-targeted test inputs using defect-guided mutation, web-search-assisted fidelity assurance, and diversity-aware seed maintenance. Experiments across 16 framework--model--dataset settings show that EMBERS achieves 3.21 times the failure-discovery throughput of the strongest baseline under identical testing budgets. Human and LLM-assisted evaluations demonstrate that most generated inputs preserve the original information need and necessary contextual constraints. Moreover, model fine-tuning experiments show consistent-answer performance improvement, highlighting their value for robustness testing and model improvement.

0.11

CHATDIFY: Automating End-to-End Dify Workflow Synthesis from Natural Language via Multi-Agent Collaboration

cs.AI ·2026-11-01 ·Tao Fan, Guoqiang Ma
benchmark
Abstract

The proliferation of Large Language Models has catalyzed powerful application development platforms like Dify, which enable users to build sophisticated LLM-powered workflows through modular node orchestration. However, manually configuring these workflows requires deep platform expertise and remains time-consuming and error-prone, creating a significant barrier for non-expert users. To address this challenge, we propose \textsc{ChatDify}, a novel multi-agent framework that automates end-to-end Dify workflow synthesis from natural language descriptions. \textsc{ChatDify} employs an Actor-Critic architecture comprising three specialized agents: the \textit{Planner Agent} that decomposes user requests into structured execution plans, the \textit{Executor Agent} that performs context-aware parameter synthesis and schema-augmented DSL compilation, and the \textit{Reviewer Agent} that ensures reliability through hybrid verification with closed-loop refinement. We evaluate \textsc{ChatDify} on a curated benchmark of 100 production-inspired tasks across three state-of-the-art LLM backends, achieving 0.96, 0.91, and 0.75 success rates with Qwen-3.7, GLM-5.1, and DeepSeek-v4, respectively. These results demonstrate its robustness and practical applicability for lowering the technical barrier to creating sophisticated LLM applications.

0.10

Local or Escalate? Post-Training Expert Routing for Real-Time Duplex Speech Models

cs.AI eess.AS ·2026-11-01 ·Ruoyu He, Jisen Li, Anthony Ji, Changyi Yang, Zetian Li +3 more
benchmark
Abstract

A full-duplex model should answer easy questions immediately, but hard questions often need stronger reasoning, which may require waiting. The challenge is to choose between these paths at the moment the model begins to speak. Recent systems learn related mode decisions through special-token generation inside the dialogue model, coupling routing to turn-taking and temporally annotated full-duplex training. We study a post-training alternative. We freeze MiniCPM-o 4.5, using only its speech input and output, and fit a logistic failure readout on hidden states already available at its native listen-to-speak transition. The resulting gate requires no additional model generation and exposes thresholds that can be retuned without changing the duplex policy. A probe fitted on text-input states also reads the states of spoken questions, including unseen human recordings. On the internal benchmark, selective handoff improves answer accuracy by 27.1 points. Compared with always using the expert, it reduces mean server time to first answer audio by 31.5 percent. It also outperforms matched random routing across external speech question-answering pools, as well as a zero-shot GPT-5.4 mini router that rates the transcribed question. The training recipe also transfers to a second duplex model family. These results show that an existing duplex model can be equipped to answer easy turns immediately and wait for stronger reasoning or fresher knowledge on the turns that need it, without further duplex training.

0.10

Test-Time Scaling for Open-Ended Reasoning at Small Model Scale

cs.LG cs.AIcs.CL ·2026-11-01 ·Adam Y. Ahmed, Abdullah G. Elafifi
benchmark
Abstract

A vote over a model's samples can only help when the samples disagree. For open-ended generation the binding question is therefore not how to combine samples but whether a model produces samples worth combining, a question test-time scaling has answered for math but not for free-form text. We study it at small scale with semantic cluster voting, a judge-free aggregator that groups a small instruct model's samples by meaning and returns the largest cluster's representative, with no reward model or self-judging. On a 250-prompt benchmark with three tiers of answer checkability, the same pattern holds across three model families (Qwen2.5-3B, Llama-3.2-3B, Gemma-2-2B) and two generation seeds: voting raises accuracy on the math control, where samples vary but stay checkable (a gain of +10 to +29 points across the three models), yet adds at most +0.30 judge points on open-ended text. On a larger 200-problem MATH-500 run with the primary model, the math gain is +8.5 points (95% CI [+4.0, +13.0]). Almost every open-ended sample falls into one meaning cluster, so there is little to select over. The collapse is consistent with instruction-tuning rather than model size, and raising decoding temperature does not recover usable diversity, because it costs more quality than it adds (net -0.64 judge points, 95% CI [-0.93, -0.37]).

0.10

Safety4DM: A Vision-Language Framework for Risk-Aware Dynamic Maps in Vehicle-Road-Cloud Collaborative Driving

cs.AI ·2026-11-01 ·Zhengshu Zhou, Shaoyu Wang, Lu Tao, Xinhui Hu, Jiaqi Gu +1 more
benchmark
Abstract

Dynamic map (DM) is a core information infrastructure for vehicle-road-cloud (VRC) cooperative autonomous driving. It fuses perception data from multiple sources to overcome the inherent line-of-sight and field-of-view limits of single-vehicle perception. However, DM still falls short on driving safety in two ways. First, DM data specifications already define road conditions semantically, but no pipeline yet turns real-world road observations into fine-grained risk entities in the DM. Second, existing DM systems lack a natural language interface for driving risks. To close these gaps, we propose Safety4DM, a safety-oriented language system for VRC dynamic maps. We first build RoadRisk, a real-world road image dataset that covers four types of road conditions and four risk levels. On this dataset, we fine-tune a vision-language model (VLM) for road safety risk perception. The model outputs structured risk representations that the DM can ingest directly. We then introduce a Safety-aware Step-wise Prompting (SafeStep) mechanism as the natural language query interface of the DM. SafeStep answers queries accurately and also generates actionable driving advice. To build and evaluate this interface, we develop SafeVRC-Sim, a VRC cooperative simulation framework, and SafeDM-QA, a safety question-answering benchmark. Experiments show that even the compact Qwen3-VL-4B reaches 89.83% accuracy on risk perception after fine-tuning. SafeStep also plugs easily into multiple LLM families. With medium and large models (20B and above) from five families, it achieves over 93% natural language query accuracy in every case, which shows strong generalization.

0.09

A Hybrid Anomaly Detection Method for Multivariate Time Series

cs.LG cs.AI ·2026-11-01 ·Ali Babalhavaeji, Hui Song, Mahdi Jalili
Abstract

Multivariate time series are ubiquitous in monitoring smart systems such as power plants, smart houses, and servers. These systems use multiple sensors to collect the system's status over time. Those sensors represent the status of the whole system as one entity. Anomaly detection in such smart systems is of critical importance to prevent system failure and maintain functionality. But anomaly detection in such multivariate time series data is very challenging since this type of data includes both temporal and spatial information. Capturing both of these pieces of information is important to detect anomalies. In this paper, we propose a hybrid deep learning-based anomaly detection method, which captures both temporal and spatial information. We propose to take advantage of the feature extraction capabilities of 2d convolutional network in a variational autoencoder, alongside a prediction-based deep neural network of 2d convolutional networks and a gated recurrent unit network to capture both spatial and temporal information. Experiments on public data against state-of-the-art show the potential of our method.

0.09

How Does Chain-of-Thought Supervision Teach Models to Reason? A Theoretical Analysis

cs.LG cond-mat.dis-nnstat.ML ·2026-11-01 ·Kaito Takanami, Cengiz Pehlevan
Abstract

Training with chain-of-thought (CoT) supervision is an effective approach for teaching models multi-step reasoning procedures. At inference time, however, models may need to continue reasoning beyond the finite teacher CoT sequences observed during training, and longer reasoning can instead degrade performance, a phenomenon known as overthinking. How the amount and structure of training-time CoT give rise to such failures remains theoretically poorly understood. To study this problem, we introduce an in-context weight prediction model trained on finite gradient descent trajectories for linear regression. Using high-dimensional asymptotic analysis, we show that the success and failure modes of long-horizon reasoning undergo phase transitions depending on the context length and the total amount of training CoT. We further distinguish two mechanisms of overthinking, error accumulation and error amplification, and show that with short training contexts, increasing the training CoT budget can worsen long-horizon error before eventually stabilizing reasoning. Experiments with fully trained linear and multi-head softmax attention further show that these qualitative behaviors persist. Our results provide a theoretical account linking properties of training-time CoT to inference-time failures and offer guidance for designing teacher CoT that supports stable reasoning.

0.08

Missingness-Aware Prior-Data Fitted Network for Irregular Clinical Time-Series

cs.LG ·2026-11-01 ·Hadi Mehdizavareh, Gabriele Santangelo, Giovanna Nicora, Pietro Bosoni, Simon Lebech Cichosz +3 more
Abstract

Clinical time series are irregularly observed with laboratory measurements recorded at uneven intervals and observation patterns often reflecting patient state, clinical concern, and site-specific practice. Consequently, missingness is not merely a nuisance factor but an informative signal for prediction. We propose a Missingness-Aware Prior-Data Fitted Network (MA-PFN) for irregular clinical time-series prediction. MA-PFN first learns a diffusion-based prior over joint laboratory value and observation-mask trajectories from source-domain electronic health records, capturing both physiological dynamics and measurement processes. Synthetic trajectories sampled from this prior are then transformed into context-query tasks to train a PFN-style conditional predictor. Once trained, the model performs prediction for new admissions in a single forward pass, requiring neither retraining nor parameter adaptation on target datasets. The framework naturally supports both imputation and forecasting through different context-query constructions. We conduct internal validation on a held-out MIMIC-III test set and external validation on eICU, evaluating both imputation and forecasting and assessing cross-domain generalization under shifts in patient populations, clinical practices, and observation mechanisms. MA-PFN achieves the best performance across all evaluated settings, with Overall/Macro nRMSD ranging from (0.193/0.186) for internal imputation to (0.297/0.297) for external forecasting, without target-domain adaptation. These results demonstrate that learning missingness-aware task priors enables accurate and transferable clinical time-series prediction under informative missingness.

0.08

FOCUS: Closed-Loop Attention Feedback for Efficient Vision-Language Understanding

cs.AI cs.CV ·2026-11-01 ·Akul Santhosh, Aastha Jhunjhunwala
Abstract

Processing every video frame through Vision-Language Models (VLMs) adds significant inference cost and limits their ability to comprehend real-time video. Existing approaches either sub-sample frames uniformly, ignoring content and missing important events, or prune visual tokens inside the VLM to cut per-call cost while still invoking the model on every sampled frame. We introduce FOCUS, a training-free three-layer framework that reduces the number of invocations and steers each one towards what matters. Temporal gating measures inter-frame dissimilarity in the patch-embedding space of a smaller VLM from the same family, triggering burst captures only when the scene changes in a semantically meaningful way. Patch-level dissimilarity is then accumulated across the burst and fused with a top-down attention prior into a spatial focus map that encodes where motion and semantic relevance coincide. That map calibrates the VLM's value cache during generation, progressively amplifying vision tokens in task-relevant regions and attenuating off-target ones, without removing any token from the context. The loop closes after each response: text-to-vision cross-attention is extracted from the decoder and fed back as the prior for the next burst, letting the system converge on semantically important regions over successive observations. We evaluate FOCUS on the 500-video EgoSchema subset in streaming mode, with a single Qwen3-VL-8B backbone shared by every method. It reaches 74.8% accuracy using 25.4 VLM calls per clip, against 74.6% with 55 calls for uniform sampling and 74.2% with 55 calls for BOLT. On egocentric video captioning across four domains, it matches the best baseline's quality while issuing 57% fewer VLM invocations than uniform sampling.

0.08

Beyond Multimodal Emotion Conflict: Temporal Context-Aware Hidden Emotion Detection with Multimodal Large Language Models

cs.AI ·2026-11-01 ·Yonghao Wang, Jiaxin Zhou
benchmark
Abstract · 6 pages, 4 figures

Multimodal emotion recognition commonly combines textual, acoustic, and visual signals to infer affective states. Recent work has begun to study conflicts among modalities, but a conflict does not necessarily indicate that a person is concealing an emotion. It may also result from mixed emotions, individual expressive styles, or contextual factors. This proposal aims to distinguish ordinary emotional inconsistency from emotion concealment and to infer the underlying emotion when supported by evidence. We plan to construct a multimodal benchmark with explicit annotations of expressed emotion, internal emotion, and concealment conditions, and to develop an MLLM-based framework that integrates fine-grained facial Action Units (AUs), cross-modal features, and temporal context.

0.07

Inner-Speech Decoding from EEG Across Three Corpora Under a Fixed Architecture and Protocol

cs.AI cs.LGeess.SP ·2026-11-01 ·Md Wahiduzzaman Suva, Md Imtiaj Alam Sajin, Esm E Moula Chowdhury Abha, Susham Moula Choudhury Akash, Md Tauhidul Islam +3 more
benchmarkdataset
Abstract

Inner-speech decoding from electroencephalography could support non-invasive communication for people with severe motor impairment, but comparison across studies is hindered by dataset- and protocol-specific evaluation. We present a controlled cross-dataset study using one preprocessing pipeline, one within-subject five-fold cross-validation protocol, and a compact dual-attention Transformer (VC-DATNet; approximately 131,000 parameters) across three public corpora: Thinking Out Loud (10 participants), BISpeech (4), and SyncSpeech (3). VC-DATNet achieved per-participant sensor/source hybrid accuracies of 39.48% on the four-class Thinking Out Loud task, 20.60% on the eight-word BISpeech task, and 25.53% on the eight-word SyncSpeech task, for which we report the first decoding result. Sensor-space and cortical source-space representations were complementary, with each performing better for different participants; this pattern replicated across all three corpora and nine architectures. Moreover, ten decoders spanning convolutional, recurrent, temporal-convolutional, hybrid Transformer, and pretrained foundation-model families converged within a narrow accuracy range on every dataset. These findings indicate that representation choice should be individualized and suggest that performance on these public corpora is constrained more by signal quality and dataset scale than by model capacity. The released pipeline provides a reproducible benchmark for future inner-speech decoding studies

0.07

ReFrame: Silence-Aware Compute Reclamation for Full-Duplex Speech Models

cs.LG ·2026-11-01 ·Zheng Zhou, Guanyi Chen, Jing Wei, Haoyuan Liang, Jian Zhang +1 more
Abstract · Corrected citations for Moshika-RL and Full-Duplex-Bench

Full-duplex speech models execute dense model inference for every audio frame throughout a live interaction. This uniform framewise schedule creates sustained GPU load across dialogue states whose computation has strong, reusable temporal structure. ReFrame formulates full-duplex inference as a silence-aware computation problem. Mutual Silence Speculation (MSS) operates when both participants are silent: it evaluates multi-frame Temporal continuations, verifies text and hidden-state stability, and applies batched Mimi audio-token correction. Listening Frame Folding (LFF) operates while the user speaks and the agent listens: it processes already-arrived user frames through parallel Temporal execution and frame-batched audio-token generation while accepting only the verified prefix. Frame-Budget Reinvestment (FBR) assigns the released frame intervals to resumable external retrieval and language-model workers, then injects timely references through Moshi's conditioning interface. We instantiate ReFrame on five full-duplex models spanning Moshi-family, phase-structured, and flattened autoregressive generation. Across the complete 1,225-conversation FDBench, ReFrame reduces measured GPU execution time by 18.80-31.01% and retains 96.92-100% non-PAD text agreement on five different models. On HaluEval tests, FBR increases the number of correct generated spoken answers from 86 to 205 under the same auxiliary-service budget. ReFrame turns recurring dialogue structure into lower inference cost and higher end-to-end service capability.

0.07

Large-Scale Bayesian Optimisation Using Uncertainty-Calibrated Product-of-Experts Gaussian Process Models with Trust Regions

cs.LG ·2026-11-01 ·Yean Hoon Ong
Abstract

The use of collections of independent local Gaussian processes (GP-ind) combined with trust regions is a widely adopted approach for large-scale and high-dimensional Bayesian optimisation (BO). However, this approach suffers from several limitations, including discontinuities at the boundaries of disjoint regions, an inability to capture long-range spatial correlations, potential information loss, and abrupt increases in computational overhead. To address these limitations, this paper introduces a novel BO algorithm, termed BO-pro-c-tr. The proposed method employs a product-of-experts Gaussian process model with uncertainty calibration (GP-pro-c) as its surrogate model and incorporates a one-to-many association strategy for integrating GP-pro-c with trust regions. Empirical results suggest that, when combined with the GP-LCB acquisition function, BO-pro-c-tr achieves an average reduction of 2.6% in simple regret and 34.9% in computational overhead relative to BO algorithms based on independent local GP models coupled with trust regions.

0.07

Seeing What Matters, Verifying What Pays: Load-Adaptive Speculative Decoding for Vision-Language Models

cs.AI cs.CV ·2026-11-01 ·Hong Nie, Zhangchi Gao, Yuqian Liang, Jun Yu, Bo Wang +2 more
Abstract

Speculative decoding accelerates vision-language model (VLM) inference without altering the output distribution, yet two factors limit its gains. Existing drafters either lack the image evidence required for visually grounded tokens or pay to condition on every visual token, although a short draft block typically depends on only a few image regions. Moreover, the value of verifying additional candidates varies with GPU load, so no fixed verification structure is efficient across serving conditions. We present Uni-LaSpec, a block-parallel speculative decoding framework that addresses both factors. To improve draft quality, Uni-LaSpec reuses the verifier's own attention: the query retained from the preceding verification ranks the cached, int8-quantized visual keys, and a cross-layer vote selects a few visual tokens that the drafter fuses through gated cross-attention. Acceptance-aware training further weights each draft position by its contribution to the accepted prefix. To allocate verification, a marginal-cost criterion selects the structure per request and per round: draft trees exploit idle verifier capacity under light load, whereas a dynamic head truncates linear drafts under heavy load. Across six VLMs from three model families and 17 text and multimodal benchmarks, selecting eight visual tokens retains at least 98.6% of the acceptance obtained from the full image at sub-millisecond selection cost. With greedy decoding at concurrency one, Uni-LaSpec achieves up to 5.11x text and 3.24x multimodal throughput over autoregressive decoding. Across six load levels and three GPU types, it outperforms autoregressive decoding in every setting and matches or exceeds the best fixed verification strategy in 123 of 144 settings; under bursty arrivals, it reduces mean end-to-end latency by 47.6% relative to linear verification.

0.07

Language, Vision & Media 154

Multi-criteria Rating and Review based Recommendation Model

cs.IR ·2026-11-01 ·Emrul Hasan, Chen Ding, Alfredo Cuzzocrea
publishedpositionhighly cited
Abstract · 10 pages. Accepted version of the paper published in the 2022 IEEE International Conference on Big Data (Big Data). (c) 2022 IEEE

These days, due to the advancement of information technology, recommendation system has become one of the key tools for e-commerce business. E-commerce platforms allow users to provide feedback in both written comments and numerical ratings. Recommendation systems are utilized to recommend users to new or unseen items based on these previously collected comments or ratings. In recent years, multi-aspect or multi-criteria based recommendation systems have been studied a lot by the recommendation research community. However, these research works are conducted either with reviews or ratings, not with both. In this project, we argue that integrating both textual reviews (with multiple aspects) and numerical multi-criteria ratings can further enhance the overall rating prediction accuracy. We propose a Multi-criteria Rating and Review based Recommendation model (MRRRec). We show that incorporating multi-criteria ratings into multi-aspect ratings from reviews has a great impact on performance. Our proposed model outperforms several state-of-the-art models such as ANR, DeepCoNN, and Deep Multi-criteria Recommendation System in terms of MSE, MAE, precision, recall, and F1. We show that our proposed model achieves an average of 19% and 23.0% lower MSE and MAE respectively and 7.0%, 1.0% and 3.8% higher precision, recall, and F1 score respectively. We further show that our model performs significantly better with Word2Vec word embedding than the GloVe word embedding method.

0.68

TopoSteer: Topological Layer Selection for Activation Steering in LLMs

cs.CL cs.LG ·2026-11-01 ·Tuc Nguyen, Marcos Gutiérrez-del-Pozo, Eduardo Paluzo-Hidalgo, Yifan Hu, Adaku Uchendu +1 more
position
Abstract · 42 pages, 8 figures

Activation steering controls large language models at inference time by adding a vector to their internal representations. Most work asks how to build that vector, and far less attention has gone to where to add it, even though steering at a poorly chosen layer can be worse than not steering at all. Prior selectors choose a layer without ever steering the model to test it: they score how separable the target behavior is at that layer, and how consistently the individual examples point along the steering direction. We argue that what matters is not the separability of the two activation clouds, nor the alignment of individual displacements with their mean, but the extent of the cloud of per-example displacements that mean-difference steering averages. We propose a topology-based layer selection method, called TopoSteer, which scores a layer by the total H_0 persistence of its displacement cloud, computed as the length of the cloud's Euclidean minimum spanning tree. We show that the score focuses on the displacement cloud's relative geometry, while class separation affects it only through standardization. We also provide a theoretical explanation for our method by proving that, as example-level variation increases, TopoSteer and discriminability move in opposite directions, so the two criteria can rank layers in opposite ways. Across multiple language models and steering tasks, TopoSteer provides a principled alternative to discriminability- and separability-based layer selection and yields more effective steering without requiring held-out validation.

0.59

Adversarial Robustness of Retrieval-Based Fact-Checking: Detecting and Mitigating Evidence Contamination

cs.IR cs.LG ·2026-11-01 ·Alamgir Munir Qazi, John P. McCrae, Jamal Abdul Nasir
top venuecode/data
Abstract · Published in Findings of the Association for Computational Linguistics: EMNLP 2026. 14 pages, 5 figures, 7 tables. Code and dataset: https:/…

Retrieval-based fact-checking systems ground verification in large document collections, creating a critical vulnerability: adversarial evidence contamination. We demonstrate that factually inconsistent but contextually coherent evidence generated to satisfy both retrieval-similarity and linguistic-fluency constraints can systematically mislead verification models. Contaminating just 3.1-7.4% of training corpora with five label-specific strategies causes adversarial content to penetrate up to 35.2% of top-k retrievals, degrading verification accuracy by up to 46.9% relative F1. Existing retrieval-stage defenses, validated against gradient-optimized or syntactically anomalous attacks, transfer poorly to this threat model: detection rates collapse and F1 recovery exhibits high variance across seeds. We propose GLAD (Geometric-Lexical Adversarial Defense), which fuses dense embedding peer-coherence with lexical distributional isolation. Across two benchmarks, two embedding models, and five random seeds, GLAD achieves the highest detection rates (70-86%) and the most consistent F1 recovery, reclaiming 73-80% of contamination-induced degradation on LIAR-RAW while approaching clean baselines on RAWFC. Crucially, unlike defenses that trade clean performance for robustness, GLAD imposes negligible cost on uncontaminated corpora.

0.57

EarlySelectKV: Anticipatory KV Selection via Cross-Layer Query Proxies

cs.CL ·2026-11-01 ·Yuhao Liu, Yuehao Li, Linghe Zhang, Gaobin Huang, Junran Li +2 more
top venuecode/data
Abstract · Published in EMNLP 2026. 20 pages. Code: https://github.com/sea-with-sakura/EarlySelectKV

Query-conditioned sparse KV selection reduces long-context inference cost by attending to a subset of historical KV entries at each decoding step. Existing selectors must wait for the target-layer query before scoring candidates and generating indices, placing routing on the immediate pre-attention path. We introduce EarlySelectKV, a training-free transformation that uses proxy queries from earlier hidden states to precompute selected KV indices while retaining the real target-layer query for final attention. Tests with RocketKV, Quest, and Loki show no consistent pattern of quality degradation across the evaluated settings. In the primary RocketKV evaluation, ES-RocketKV differs by +0.018 points on LongBench, approximately -0.01 points on RULER, and +0.25 percentage points on representative NIAH/Passkey settings. In a TensorRT-LLM deployment with CPU-offloaded routing, ES-RocketKV improves end-to-end output throughput by up to 12.4% and reduces GPU-resident memory by 11.1% of the combined KV/K-summary pool. This makes routing timing an explicit optimization target alongside KV subset selection in sparse long-context decoding. Code is available at https://github.com/sea-with-sakura/EarlySelectKV.

0.57

Object-Only Is Not Context-Free: A Counterfactual Diagnostic for Context Removal in Object Retrieval

cs.CV ·2026-11-01 ·Boaz Meivar, Shai Avidan
publishedcode/data
Abstract · Accepted at WACV 2027. 8 pages plus references and supplementary material (15 pages total). Code and adapter checkpoints: https://github.com…

Object-centric retrieval pipelines often mask the background before pooling backbone features, aiming for a descriptor that reflects the object rather than its scene. Whether this truly yields an object-only descriptor, however, is rarely tested. We introduce the Keyhole Aperture Probe (KAP), a counterfactual diagnostic that repurposes the ground-truth masks of existing panoptic segmentation datasets (COCO, ADE20K) for retrieval: it holds the pooled object region fixed and varies only the pixels outside it, summarizing a backbone's context dependence in a single number. We find that the pooled object region has already integrated its surround. Replacing the context, with a neutral fill or even a different real scene, changes the descriptor, and when retrieval fails under a blank fill, the descriptor matches plain-background images rather than its own object. Object-only is not context-free. Across ten modern backbones, object-only descriptors retain only about a quarter to two-thirds of their full-image accuracy, and the effect is not unique to Transformer self-attention: a convolutional backbone shows it too. The dependence is also reducible: a lightweight adapter that encourages the descriptor to stay stable as context is removed lowers it and improves full-image instance retrieval on INSTRE, transferring to ILIAS with accuracy improved or preserved. A matched contrastive-fine-tuning control does not reproduce those gains. We release KAP as a simple, backbone-agnostic evaluation protocol to measure and reduce this dependence before deployment.

0.52

MedPsy: State-of-the-Art Small Medical Language Models for Efficient Edge Deployment

cs.CL ·2026-11-01 ·Davide Vitabile, Alexandro Buffa, Akshay Nambiar, Amril Nazir
top venuecode/data
Abstract · Accepted at NeurIPS 2026 (main conference)

Medical language models promise expert clinical support, but the strongest open-source LLMs remain too large for private edge-device deployment, while compact medical models still trail substantially on knowledge-intensive and real-world healthcare tasks. Reasoning models that fit edge memory budgets can produce long reasoning traces, increasing sequential decoding, end-to-end latency, energy consumption, and KV-cache memory pressure. We present MedPsy, a family of text-only 1.7B and 4B medical language models designed for edge deployment, targeting this accuracy-generation-efficiency trade-off. Our recipe combines a synthetic medical data pipeline over biology, medicine, and health seeds with chain-of-thought targets from a 235B medical-focused LLM teacher; a four-stage post-training curriculum comprising two SFT stages followed by two RL stages with hard-sample mining; and an edge deployment evaluation spanning multiple GGUF quantization variants and real-device inference on laptop and smartphone hardware. On seven closed-ended medical benchmarks, MedPsy-4B surpasses MedGemma-1.5-4B by 19.34 points and matches MedGemma-27B while being 6.75x smaller; MedPsy-1.7B outperforms MedGemma-1.5-4B by 11.42 points despite being less than half its size. On HealthBench-Hard, MedPsy-4B surpasses MedGemma-27B by 15.33 points, while MedPsy-1.7B surpasses it by 11.66 points despite being ~16x smaller. Beyond accuracy, MedPsy reduces average response length by 1.7x (1.7B) and 3.2x (4B) versus its Qwen3 backbones, while Q4_K_M quantization preserves the aggregate score within 1 point of BF16 with ~69% lower disk footprint. In real-device inference, MedPsy-4B's shorter generations translate into 75% lower median end-to-end latency and estimated energy per answer than its Qwen3 backbone at matched quantization, while peak RAM remains below 5 GB across our tested laptop and smartphone configurations.

0.51

Multicriteria Recommendation System by Leveraging Predefined, Implicit, and Undefined Criteria

cs.IR ·2026-11-01 ·Emrul Hasan, Chen Ding
published
Abstract · 9 pages. Preprint (submitted version, prior to peer review). The Version of Record is published in Social Networks Analysis and Mining (ASON…

Data sparsity, a long-standing problem in the recommendation system, occurs when the user-item utility matrix is sparse. For years, many approaches have been experimented with, including utilizing user's textual reviews and multiple criteria ratings. In a review-based recommendation system, user preferences and item attributes are learned from reviews, while in multi-criteria recommendation systems, criteria ratings are used for the recommendation task. Multi-criteria are domain-specific pre-defined explicit criteria and users are bound to provide feedback only on these criteria with discrete values. On the contrary, users are free to express their opinions through textual reviews, which can cover multiple perspectives or aspects. It is possible that these pre-defined, domain-specific explicit criteria are not enough to fully capture the multi-faceted user preferences. To this end, we propose a novel multi-criteria recommendation model that utilizes predefined, implicit, and undefined criteria. We use a semantic similarity-based sentence clustering method to identify the predefined and implicit criteria and a sentiment analyzer to estimate their ratings. The criteria that are extracted from the review and are aligned with predefined criteria are referred to as implicit criteria. A sentence is considered as expressing opinions on an undefined criterion if the similarity score between this sentence and all the predefined criteria is lower than a predefined threshold. Ratings are computed for each extracted implicit criterion and the undefined criterion based on the review content. Finally, we use all three types of criteria and an aggregation model to make the final rating prediction for the recommendation system. Our proposed method demonstrates the superiority compared to several baselines on TripAdvisor and Beer Advocate datasets.

0.49

Count-Density Networks for Modal Plate Parameter Estimation

eess.AS ·2026-11-01 ·Rodrigo Diaz, Pablo Tablas De Paula, David Marttila, Ilias Ibnyahya, Chin-Yun Yu
publishedcode/data
Abstract · 3 pages, 1 figure, 1 table. DAFx26 Parameter Estimation Challenge, Task B. Code: https://github.com/davidmarttila/dafx_challenge

We describe two submissions to Task B of the 1st DAFx Parameter Estimation Challenge, which estimates an unknown number of modal frequency, decay, and gain triples from a synthetic plate-reverb impulse response. The first method combines pooled spectral features with time-domain and absolute-scale conditioning in a real-valued convolutional count-density network, while the second uses a complex-valued Transformer count-density network. Both methods jointly infer the modal count and per-mode attributes directly from the IR. On an independently generated 100-IR comparison set, the two neural estimators achieve lower overall challenge error than the evaluated classical baselines, with frequency and decay estimation substantially more accurate than gain estimation.

0.48

OpenULF: A Modular Toolkit Enables Data-Driven Ultra-Low-Field MRI with both Classical and Quantum Sensors

eess.IV cs.CVphysics.med-ph ·2026-11-01 ·Hanhan Wu, Yuchen Yuan, Weiwen Jiang
publishedcode/data
Abstract · 8 pages, 6 figures, 3 tables. Accepted for publication in the 44th IEEE International Conference on Computer Design (ICCD 2026). Code availa…

Ultra-low-field (ULF) Magnetic Resonance Imaging (MRI) offers a portable, low-cost alternative to conventional scanners by operating at field strengths below 10 mT. At these fields, two sensor technologies compete: quantum SQUID (Superconducting Quantum Interference Device) sensors, which provide high sensitivity but require cryogenic cooling, and classical induction coils, which are inexpensive and room-temperature but produce weaker signals. With conventional physics-based imaging, both suffer from low signal-to-noise ratios (SNR). Data-driven reconstruction has overcome similar limits in related inverse problems, but its use for ULF-MRI has been hindered by a lack of open datasets, benchmarks, and flexible simulators. To close this gap, we present OpenULF, a modular, open-source simulation toolkit that generates paired training data and supports end-to-end machine-learning (ML) reconstruction. OpenULF integrates a Bloch-equation-validated physics simulator, calibrated SQUID and coil receiver models, reconstruction ML networks, and an automated design-space exploration engine. Architecturally, a receiver abstract base class, a typed HDF5 contract that decouples physics from learning, and a resumable sweep engine make sensors, models, and datasets independently substitutable. Using OpenULF, we train 774 models across four anatomical datasets and conduct three design-space explorations. The effectiveness of ML reconstruction depends strongly on the dataset and the sensor. We identify a SQUID-to-coil crossover near 450 $μ$T and demonstrate that sensor hardware and ML model capacity can be traded off during system design. Sensor and reconstruction choices must therefore be co-designed: the best sensor is set by the dataset and field strength, not by SNR.

0.48

Switch-SWE: Do Coding Agents Hold Focus When Switching Between Unfinished Tasks?

cs.CL ·2026-11-01 ·Shuochen Chang, Qianli Ma, Yibo Miao, Qingyang Liu, Bingjie Gao +5 more
benchmarkcode/data
Abstract

Software development sessions often involve several unfinished tasks rather than one issue carried through to completion. As coding agents take on repository-level development, they need to recover an earlier task's requirements and partial progress after intervening work. Existing repository-level benchmarks typically score one issue-to-patch episode and do not directly measure whether an agent can resume a task after advancing other unfinished work in the same repository. We introduce \textbf{\ourmethod}, a benchmark for focus-task resumption in sessions that interleave unfinished tasks drawn from the same source repository, with a persistent checkout for each task. Each focus task is evaluated in matched uninterrupted and switched sessions with identical focus requirements and executable verifiers. The switched session interleaves ordered stages from other unfinished tasks in the same repository, while only the final focus patch is scored. \ourmethod~contains 200 focus tasks from 19 Python repositories, reconstructed from issue, pull-request, and review histories and rendered as 400 matched session specifications. Across 7 coding agents, interleaved unfinished work produces an average focus-task resolution gap of 11.5 percentage points, ranging from 6.0 to 15.0 percentage points. Trajectory analysis identifies cross-task requirement confusion and loss of task state as recurrent resumption behaviors. We introduce a plug-in context-management module that selects task-relevant history and maintains concise task-state summaries. On OpenCode, the module improves focus-task completion under same-repository task switching by 5.5 to 15.5 percentage points across the evaluated models. Code is available at github.com/Miraclecsc/Switch-SWE.

0.48

RadPose: ISAC-based RF 3D Human Pose Estimation

cs.CV cs.NI ·2026-11-01 ·Geo Jie Zhou, Kunzhe Song, Ruxin Lin, Huacheng Zeng
publishedcode/data
Abstract · Accepted to WACV 2027

Human pose estimation under adverse lighting and occlusion conditions remains challenging for camera- and LiDAR-based systems. Although mmWave radars provide robustness to illumination and occlusion, they rely on dedicated hardware and licensed spectrum, limiting scalability. In this paper, we present RadPose, a learning-based monostatic integrated sensing and communication (ISAC) framework that enables mmWave communication devices to perform 3D human pose estimation. Unlike specialized radars, RadPose leverages monostatic full-duplex operation to extract channel impulse responses (CIRs) directly from standard OFDM communication signals, generating 3D radio point clouds without additional sensing hardware. To address the sparsity and noise of RF observations, RadPose employs a cross-frame transformer-based pose inference network with semantic-guided spatial attention and target-driven temporal attention for robust spatio-temporal aggregation. It further adopts a joint-aware inverse kinematics refinement module based on graph convolution to enforce skeletal topology, bone-length consistency, and anatomical plausibility. Extensive experiments demonstrate RadPose achieves competitive accuracy compared to state-of-the-art radar-based approaches while maintaining robust performance in unseen and occluded scenarios. The dataset and source code are available at https://github.com/Geo-Chou/RadPose.

0.46

SEALECT: Scene-Conditioned Pose Priors for Plausible Maritime Vessel Retrieval

cs.CV cs.LG ·2026-11-01 ·Faisal Mohammad, Mohamed Younis
publishedcode/data
Abstract · Accepted to WACV 2027 (Applications Track). 22 pages: 11-page main paper including references, plus 11 pages of supplementary material. Code…

We address scene-conditioned object retrieval for compositional augmentation in maritime imagery, where strong geometric constraints make appearance-only retrieval produce physically implausible vessel insertions. We propose SEALECT, a retrieval framework that jointly models visual compatibility, vessel pose, scale, and scene geometry through a unified score combining neural similarity, scene-conditioned pose priors, and observed-pose consistency. We construct a maritime scene-pose library and model the pose and scale priors nonparametrically, as a kernel that conditions each prior continuously on the scene with no learned parameters. Across seven maritime benchmarks, our trained scene-conditioned retriever exceeds the strongest appearance encoder (SigLIP 2) in recall (+0.14), and the training-free prior raises geometric plausibility further at negligible recall cost, more than halving the constraint violations of appearance retrieval and outperforming learned foreground-search baselines (GALA-style, DiscoFOS). In a blind study of end-to-end composites (BootPlace placement, DiffHarmony harmonization), SEALECT ranks first among six methods, ahead of random and GALA-style selection and statistically tied with CLIP, DiscoFOS, and DINO.

0.46

Synthetic Realities: Evaluating Human Ability to Distinguish AIGenerated Videos from Real Footage

cs.CV cs.HC ·2026-11-01 ·Danyal Sarfraz, Raja Mubashar Karim, KwanMyung Kim
published
Abstract · Published in Proceedings of IASDR 2025

Recent advances in generative AI have enabled multiple pathways for high-fidelity video synthesis: text-to-video, image-to-video animation, and video outpainting. Empirical side-by-side evaluations of how untrained viewers perceive and distinguish these outputs from real footage remain scarce. In this study, we systematically compare human detection accuracy across these three AI generation techniques within three thematic contexts: historical footage, film and media content, and natural environments. We constructed a balanced stimulus set comprising equal amounts of real and AI-generated videos (18 each). The AI clips were evenly distributed across the three generation methods using Google's Veo 3, Lightricks LTX Video, and Wan VACE. All videos were produced and standardized within the ComfyUI framework to ensure consistent quality and duration. 87 participants judged every clip in a binary forced-choice task ("real" vs. "AI-generated"). Participants correctly identified videos 60% of the time on average. Image-to-video clips were recognized most accurately (79%), followed by real footage (64%), outpainting (49%), and text-to-video (43%). Accuracy also varied by theme: film and historical scenes yielded higher detection rates than environmental clips, which were frequently mistaken for AI. Logistic regression confirmed significant effects of both technique and theme as well as their interaction (p < 0.001), indicating that detection success depends jointly on how the content was generated and what it depicts. Findings reveal a consistent bias toward assuming synthetic origins and highlight that perceptual realism in AI video is shaped more by context than by model type, underscoring the importance of media-literacy approaches and context-aware evaluation tools for navigating increasingly synthetic visual media.

0.45

Current state of LLMs for Arabic dialectal machine translation

cs.CL ·2026-11-01 ·Josef Jon, Rawan Bondok, Ondřej Bojar
highly cited
Abstract

This work presents an evaluation of large language models (LLMs) for English to dialectal Arabic machine translation on the MADAR dataset. We evaluate both translation directions (English to Arabic and vice-versa) on 16 Arabic dialects. Our experiments cover a diverse set of models, including specialized Arabic models (Jais, Nile), multilingual models (Gemma, Command-R, Mistral, Aya), and commercial APIs (GPT-4.1). We employ multiple evaluation metrics: BLEU, CHRF, COMET (both reference-based and reference-less variants) and GEMBA (LLM-as-a-judge), as well as a small-scale manual evaluation, to assess translation quality. We discuss the challenges of automatic MT evaluation, especially in the context of Arabic dialects. We also evaluate the ability of LLMs to classify the dialect used in a text. The study offers insights into the capabilities and limitations of current LLMs for dialectal Arabic machine translation, particularly highlighting the difficulty of handling dialectal diversity, although the results may be influenced by possible training data contamination, which is always a concern with LLMs.

0.45

Don't Guess, Ask Wisely: Benchmarking and Training Proactive Elicitation in Open-Domain LLM Conversations

cs.CL ·2026-11-01 ·Rongyu Zhang, Mutian Bao, Jinjian Zhang, Yuetian Du, He Xu +6 more
top venuebenchmark
Abstract · 26 pages, 6 figures; EMNLP 2026 oral

When user queries are incomplete, large language models should ask targeted questions before committing to an answer, a capability we call proactive elicitation, yet evaluation on real user-assistant queries across seven LLMs (8B to frontier) shows that fewer than 11% of underspecified queries trigger a follow-up question. To benchmark and train this capability, we introduce VagueBench, where each seed query is annotated with structured intent and detail units then degraded into intent-vague and fully-vague variants via controlled removal of linguistic cues, yielding known hidden information sets for evaluating both when and what to ask. We propose DialogGRPO, an offline reinforcement learning method that organizes multi-turn elicitation as a dialogue-state directed acyclic graph, where candidate questions at each node receive dense turn-level rewards combining immediate information gain with lookahead value through group-relative comparison, enabling fine-grained credit assignment without runtime simulation. DialogGRPO elicits missing information on over 88% of vague queries with substantial gains in both recovery efficiency and answer coverage, without over-asking on complete queries, outperforms prompting and test-time-scaling baselines, and generalizes zero-shot to MultiWOZ 2.4 and AmbigQA.

0.44

GatedV: Self-Gated Values for Transformer Language Models

cs.CL cs.LG ·2026-11-01 ·Li-Ming Yu, Hongyu Zhou
benchmarkcode/data
Abstract · 12 pages, 3 figures. Added Hongyu Zhou as a co-author. Code and experimental results are available at https://github.com/yuliming1996/GatedV

Standard attention forms query (Q), key (K), and value (V) matrices through learned linear projections of input hidden states. Architectures such as GAU and Mega incorporate nonlinear transformations into attention. In this paper, we propose GatedV, a simple attention variant that applies SiLU only to V before aggregation without introducing additional trainable parameters. We evaluate GatedV against a standard decoder-only Transformer using matched parameter counts, paired initializations, and equal training-token budgets. At 2.33M parameters, GatedV reduces mean held-out perplexity from 59.1 to 57.9 (2.0%), with improvements across all five paired seeds. At 26.16M parameters, following an equal-budget learning-rate search, mean held-out perplexity falls from 20.2 to 19.7 (2.4%), with improvements across all three paired seeds. Small-model ablations show that the tested affine rescaling of values does not fully reproduce GatedV's gains. A short interleaved benchmark on NVIDIA DGX Spark measures approximately 0.7% higher full-context forward latency for the larger model, based on synchronized wall-clock timings. These findings suggest that value-only self-gating can improve language modeling performance without increasing parameter count or redesigning the attention-weighting mechanism.

0.43

Decision Models as Policy Gates for Coding Agents: Explicit Policy Boundaries Improve Detection and Reduce False Alarms

cs.CL cs.CRcs.SE ·2026-11-01 ·James P. Beno
datasetcode/data
Abstract · 15 pages, 2 figures. Dataset, rule texts and harness available at https://github.com/agentapprove/decision-policy-rules

Coding agents can destroy data, leak secrets and message the wrong person faster than a human can review. Could a decision model catch these risks at runtime? We evaluated 22 hosted decision models and two prompted general models as classifiers for a multi-rule policy engine. A synthetic dataset of 2,643 tool calls was labeled blind against 28 rules, with a sealed holdout. With a single shared preamble in both conditions, explicit boundaries written as examples and counterexamples raised recall by 5.5-26.1 percentage points on all 23 available models and reduced false positives on 21. Jev 1.14's recall rose from 89.3% to 98.2%, while unnecessary interruptions fell from 30.6% to 4.8% of 252 clean actions. With examples, its median latency was 120 ms at $0.46 per thousand requests. Strong models retained the recall gains when concrete names were replaced by generic descriptions; three largely retained them when the same boundaries were written as prose. On 16 new scenarios written after the initial studies, a follow-up improved recall on 22 of 23 models and reduced unnecessary interruptions on 21. We release the dataset, rule texts and harness.

0.43

TAC-CoT: Tool-Empowered Implicit Chain-of-Thought via Interleaving Explicit Tokens

cs.CL ·2026-11-01 ·Juseon Do, Sangmin Bae, Jeesu Jung, Hwanjun Song
top venue
Abstract · EMNLP 2026 Main Long

LLM-based agentic systems are increasingly combined with CoT reasoning to solve tasks that require accurate reasoning, external knowledge, and multi-step deliberation. Yet, explicit CoT incurs substantial generation overhead, while implicit CoT, which replaces textual rationales with compact continuous tokens, does not naturally support the explicit communication interface required for agentic tool-use. We propose TAC-CoT, a tool-empowered adaptive implicit CoT framework that replaces explicit reasoning traces with compact latent reasoning segments while retaining textual tokens only for tool calls, observations, and final answers. This interleaved design enables agentic tool-use without verbalizing every intermediate reasoning step. Experiments on agentic math and QA benchmarks show that TAC-CoT improves accuracy by up to 8.8 points while reducing model forward passes by up to 6.8x. Furthermore, analyses show that TAC-CoT improves tool-use quality in both single- and multi-tool settings.

0.43

Testing concept-level audits of melanoma classifiers with diffusion counterfactuals and matched positive controls

cs.CV cs.AIcs.LG ·2026-11-01 ·Md Shohanoor Rahman, Md Hasibul Islam, Md Kishor Morol, Dip Nandi, Tze Hui Liew
code/dataunusual
Abstract · 25 pages, 7 figures, 9 tables. Under review at Scientific Reports. Code and data at https://github.com/srahman7900/melanoma-concept-audit

Concept-based explanations of melanoma classifiers are rarely tested against controls with known targets. We built a counterfactual audit from a diffusion model conditioned on seven 7-point checklist concepts and tested it with such controls. Colour-calibrated start noise reduced colour outliers from 78.5-84.8\% to 6.6-13.3\%. Four of seven knobs were target-dominant but also changed other concepts. Quartile-based positive controls were confounded, and the audit failed its pre-specified rule. It recovered the targets of all 16 controls matched on the generator's pseudo-labels and of 12 of 12 controls labelled by a second evaluator, for three concepts; concept activation vectors recovered fewer (11 of 16). In 18 permuted-label classifiers, significant sweep responses and target rankings also arose, whereas standardised effect margins separated the controls from these null classifiers. The pigmentation sweep raised the melanoma logit in 15 of 15 ISIC 2020 classifiers (+3.12 on average) and 6 of 6 HAM10000 classifiers (+4.01), about four times the null response in scale-free units; the regression effect did not transfer. An exploratory decomposition, evaluated against the controls, suggested that knob cross-talk partly hides vascular responses. Matched controls and null classifiers are practical checks of concept audits.

0.42

Resolve, Clear, or Defer: Training-Free Active Prototype Triage for Ecological Detection Pipelines

cs.CV ·2026-11-01 ·Anirudh Potlapally, Brijesh Nanda Brunda Yogananda, Jianyang Gu, Maanav Jajodia, Luke Meyers +2 more
published
Abstract · Accepted to the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2027), Main Conference. 27 pages including supplementary…

Ecological monitoring increasingly relies on camera traps and aerial surveys that generate large image collections. High-recall animal detectors can localize candidate animals at scale. However, they also produce many detections that still require expert review, including false positives, ambiguous cases, and animals difficult to identify at the species level. Since exhaustive review requires substantial effort, we propose training-free active prototype triage, where reviewed detections become reusable evidence to (1) Resolve supported foreground classes, (2) Clear likely background false positives, or (3) Defer insufficiently supported cases for expert review. Across camera-trap and aerial datasets, foreground recognition and false-positive rejection benefit from different pretrained representations: BioCLIP2 provides foreground evidence while DINOv3 provides background evidence, reducing erroneous foreground clearing while improving automatic foreground recovery. Deferring unsupported cases avoids the higher error rates caused by forced decisions. Compared with confidence-thresholded multiclass k-NN and uncertainty-sampling logistic regression, prototype triage improves foreground recovery while maintaining competitive or lower review burden. Acquisition goes beyond class discovery: even after classes are represented, prototype choice determines how reliably evidence propagates through the remaining detections. Our results suggest that once animal localization reaches high recall, the challenge shifts from finding animals to deciding which examples to review, how to reuse them, and when evidence does not support an automatic decision.

0.42

BanglishDepYT: Depression Detection in Native Bengali Social Media with Emotion-Preserving Preprocessing and Domain-Adaptive Pretraining

cs.CL cs.CYcs.LG ·2026-11-01 ·Ayon Roy, Sadman Sharif, Syed Shams Islam
datasetcode/data
Abstract · 39 pages

Depression affects millions of Bengali speakers across South Asia, yet automated detection systems have relied almost entirely on datasets translated from English. Such datasets cannot capture the emoticons, romanized Bengali words, and elongated spellings that characterize real expressions of distress. We introduce BanglishDepYT, a Bengali depression dataset collected natively from YouTube, comprising 191,000 unlabeled comments and 1,000 samples verified through dual bilingual annotation. We propose EPH (Emotion-Preserving Hybrid preprocessing), which converts emoticons to semantic tags, preserves Banglish tokens, protects negation words, and normalizes emotional elongation. A key finding is that preprocessing choice substantially affects performance on informal text: standard cleaning reduces the Banglish Survival Rate to zero (SVM F1: 0.7062), while EPH preserves signals (CESS: 0.9996 vs. 0.6694) and raises SVM F1 to 0.8570, an improvement of 0.1508. We further apply domain-adaptive pretraining (DAPT), continuing SahajBERT pretraining on unlabeled in-domain comments. DAPT-SahajBERT achieves the highest mean F1 (0.8161) and MCC (0.6272), significantly outperforming the baseline (F1: 0.7805, p = 0.0013, d = 1.450) and matching MuRIL (p = 0.589). These results suggest that targeted domain pretraining is a competitive alternative to large-scale multilingual pretraining. LLM-assisted annotation using Qwen2.5-7B achieves substantial human agreement (kappa = 0.611), enabling pseudo-labeling of 96,071 comments. Semi-supervised training improves AUC from 0.8499 to 0.8731 (p = 0.044), offering a scalable path to expanding annotation. All code, models, and datasets are publicly released to support reproducible research.

0.41

TopoDPO: Dense Preference Alignment for Structure-Sensitive Aerial Referring Segmentation

cs.CV cs.AI ·2026-11-01 ·Huaiyuan Peng, Jisheng Dang, Wencan Zhang, Jingwen Zhao, Hong Peng +4 more
code/data
Abstract · 10 pages, 5 figures

Preference optimization for multimodal segmentation faces a granularity mismatch: dense masks contain spatially heterogeneous preference signals, whereas image-level objectives assign one shared nonlinear weight to all spatial margins. We introduce TopoDPO, a topology-aware dense preference alignment framework for structure-sensitive aerial referring segmentation. Topo-Prior mines informative preference pairs using connected-component discrepancy and macro-corner recall, while the dense objective applies the preference penalty to each local margin before spatial aggregation against a frozen supervised reference. On Topo-RS-Hard, a 4590-instance structural stress set, TopoDPO improves intersection over union (IoU) from 50.91% to 51.24% over image-level direct preference optimization (DPO) and reduces Betti-0 error from 1.34 to 0.90; paired image-cluster bootstrap intervals against image-level DPO exclude zero for all five reported metrics. Matched supervised controls and a complete factorial analysis separate the effects of topology-aware pair mining, localized correction, and dense reference-relative preference weighting. An unfiltered stratification over all 39,244 validation records further shows the clearest structural gains on samples with moderate or severe pre-alignment component-count errors. The code is publicly available at: https://github.com/phy709/TopoDPO.

0.41

ContactCut: A Partition-Aware Contact Selection Framework for False-Merge Correction in Connectomics

cs.CV ·2026-11-01 ·Byunghyun Ko, Won-Ki Jeong, Chaoli Wang
code/dataunusual
Abstract · 5 pages, 3 figures. Submitted to the IEEE International Symposium on Biomedical Imaging (ISBI) 2027

Localized false-merge correction in connectomics requires converting an identified error into an executable native 3D partition. Still, neither intermediate correction predictions nor seeded graph solvers alone determine which native source-sink constraints should define that partition. We formulate this problem as a 3D-native partition-aware contact selection problem: selecting candidate contacts based on the complete partitions they induce. Our solution comprises two components: ContactCut (CC)-Auto, which uses learned single-contact selection, and ContactCut (CC)-Search, which enables bounded paired-contact search, generates compact alternatives, and applies learned panel ranking. On historical two-way false-merge corrections from MICrONS and FlyWire, CC-Auto improves failure-aware 3D partition (FA-3D Part.) mIoU by 5.9 and 3.4 points over the strongest graph-based controls and by 12.5 and 10.0 points over the best learned baselines. A MICrONS-trained ranker transfers to FlyWire with negligible degradation. CC-Search expands oracle reachability and prioritizes near-oracle alternatives. The code is available at https://github.com/bentake/ContactCut.

0.41

RAG-TRACE: Tracing Measurement Evidence and Diagnostic Completeness in AI-Generated ECG Reports

cs.CL ·2026-11-01 ·Elyas Irankhah, Shahed Salehzehi, Parisa Hosseini
code/data
Abstract · Code and study materials available at https://github.com/Elyasirankhah/RAG-TRACE

Purpose: Retrieval-augmented generation (RAG) can introduce electrocardiogram (ECG) criteria that lack patient-specific measurement support. This study evaluates the reliability of retrieved diagnostic criteria in AI-generated ECG reports. Methods: We analyzed 1,083 human-validated PTB-XL patient records using a library of 47 criteria cards, including 33 without numeric verification rules. Language models generated structured reports with and without retrieval. Criterion applicability was evaluated using Marquette 12SL measurements, with independent assessment using Uni-G and comparison against expert-validated diagnostic labels. Results: Keyword retrieval introduced untestable or measurement-disconfirmed criteria in 46.8% of GPT-4.1-mini reports. A deterministic verification rule withdrew 647 criterion applications, including 639 without numeric checks, while preserving 268 of 269 applications independently supported by Uni-G. Restricting retrieval to 14 testable criteria reduced the initial unsupported-criterion rate to 13.8%. Under matched instructions and an identical output schema, retrieval increased unsupported-criterion reports from 17.4% to 31.1%, while hypertrophy recall remained unchanged at 62.6%. Conclusion: These findings identify a fundamental limitation of retrieval-based clinical report generation: retrieved criteria do not necessarily provide valid patient-specific evidence. Measurement-based verification can reduce unsupported criterion applications, but diagnostic validity requires independent assessment.

0.41

Detailed Derivation of The Backpropagation Algorithm

cs.CL ·2026-11-01 ·Faris Alasmary
code/data
Abstract · 41 pages, 2 figures, 2 algorithms. Code: https://github.com/farisalasmary/neural-network-numpy

Backpropagation is the standard method for computing the gradients that train neural networks, but it is usually presented in a compact vector form that skips the intermediate steps. This tutorial derives backpropagation step by step for a fully connected feedforward network with sigmoid activations and a squared-error loss. Using a running example with three layers, we first differentiate the error with respect to a single weight by following its path through the network with the chain rule. We then show how summing over all paths leads to the error terms $δ$ and to the four general backpropagation equations in matrix form. Every term is annotated with its dimensions, so each equation can be checked for consistency. The derivation is then extended to $m$ training examples, where the weight gradients become a single matrix product and the bias term corresponds to broadcasting. It is also extended to mini-batch gradient descent, including why the gradient of a mini-batch estimates the full gradient. Batch and mini-batch training are both given as algorithms. Finally, each equation is linked to the corresponding line of a NumPy implementation trained on the AHDBase dataset of handwritten Arabic digits, and the complete code is included.

0.41

Towards Fine-grained and Robust Detection of Human-AI Co-authorship

cs.CL cs.LG ·2026-11-01 ·Ziruo Zhao, Jie Fu, Wendy Hui Wang, Zining Zhu
top venue
Abstract

The widespread adoption of large language models (LLMs) is reshaping how textual content is produced, complicating text provenance, and raising concerns for misinformation, plagiarism, and intellectual property. While prior work on AI-generated text detection has progressed, one-dimensional human-AI co-authorship scores lack signals directly relevant to ``AI involvement''. It remains unclear how adversarial paraphrasing distorts detection signals. In this work, we formulate a discrete-class text provenance detection framework. This framework allows an abstraction of realistic co-authorship scenarios into distinct degrees of AI involvement, differentiating the AI polishing on the surface level from the AI updating the discourse structures. This framework reveals the mechanisms of how adversarial paraphrasing blurs the decision boundaries for co-authorship detectors. To improve robustness of the detectors, we augment training data with adversarial paraphrases generated by the state-of-the-art paraphrasing attack model, exposing detectors to challenging meaning-preserving rewrites. Extensive experiments across multiple datasets and generation models demonstrate that adversarial paraphrasing substantially degrades detectors trained only on benign data, while adversarial data augmentation leads to consistent robustness gains and good cross-generator transferability. Overall, our findings highlight the value of discrete-class formulation and principled data augmentation for building reliable detectors of human-AI co-authorship.

0.41

MEX: The Musical Expectation Benchmark Suite

cs.SD cs.AIeess.AS ·2026-11-01 ·Mathias Rose Bjare, Augustin Bouquillard, Giorgia Cantisani, Seung-Goo Kim, Stefan Lattner +1 more
publishedbenchmarkcode/data
Abstract · 9 pages, 2 figures, 3 tables. Accepted at the 26th International Society for Music Information Retrieval Conference (ISMIR), Abu Dhabi, Unit…

We propose the Musical Expectation (MEX) benchmark suite for perceptually validating computational models of musical expectation, i.e., models that estimate music information dynamics (MID) measures of perceived musical expectations, such as surprisal or uncertainty from symbolic music or musical audio. MEX collects and extends tasks for evaluating such models previously proposed in music information retrieval and cognitive sciences using standard datasets, and evaluates whether: (1) proposed MID measures decrease with repetition due to in-context learning of the computational model, (2) peaks in such measures align with boundaries of music segments, (3) pitch surprisal and uncertainty estimated from audio are consistent with those modeled in event-based symbolic music, (4) proposed MID measures can predict electrophysiological responses to music. We stress that this benchmark suite does not provide an exhaustive computational characterization of music expectation. Rather, we envision this as an active, interdisciplinary initiative, with benchmarks evolving as our understanding of the underlying perceptual processes deepens. We provide open-access code https://github.com/musicperceptmodeling/mex to estimate surprisal and uncertainty from audio using baseline models, run the tasks on standard datasets, and include utilities for running the tasks with users' own data.

0.41

CST: Lightweight Risk Triage with Source-Grounded Evidence for Web Discussions

cs.CL cs.AIcs.IR ·2026-11-01 ·Zhihua Wang, Yifeng Gao, Yunhao Feng, Rui Pan, Baiheng Zhao +2 more
code/data
Abstract · 23 pages

Web communities rely on moderators to catch harmful messages such as insults, hate speech, and scams, but no one can read every message. Existing tools give each message a risk score, which tells moderators where to look but not what to read or how to check it. Yet meaning depends on context: the same reply, "I agree.", can support or oppose removing someone from a group. Moderators therefore need source-grounded evidence, not just scores: a few key messages kept with their original wording and replies. We present CST, a lightweight triage pipeline that first decides where to look, then what to show. It scores risk by pairing character patterns, which capture exact wording, with sentence meaning from a fixed multilingual model, so it handles several languages and runs on an ordinary CPU. It then groups related messages and picks from each group in turn, so a small reading budget covers different parts of a discussion, and links every pick back to the original thread. Across six tasks from five datasets in several languages, CST ranks harmful messages more accurately than a fully fine-tuned multilingual model and, under the same budget, finds more of them than reading by score alone. In a user study, reviewers needed at least 27% less time with CST's grouped evidence while keeping decision accuracy above 94%. CST turns risk scores into evidence that moderators can read quickly and check in context. Code and dataset reconstruction resources will be released at https://github.com/Kirito14IT/CST-MIL.

0.41

Latent Harness: Compiling Agent Behaviors into Post-Block Activation Shifts

cs.CL ·2026-11-01 ·Ran Li, Shimin Di, Zeyu Ren, Ling Yue, Jiachuan Wang +1 more
Abstract

Harness engineering and self-evolving agents improve language-model agents quickly, yet their gains live outside the model. The harness is re-sent and re-paid on every call, and the standard route to internalization is a weight update that is costly to iterate and fixed once merged. We propose compiling the harness instead. Latent Harness distills an external condition into a rank-8 activation shift. The shift is applied at the output of every decoder block of a frozen backbone, at 0.03-0.05% of parameters. The condition is either an explicit process-level behavior instruction or the behavioral contract of an agentic harness. At inference the compiled bundle carries no harness text and adds no tokens. It exposes three runtime controls for injection strength, injection phase, and shift-space mixing. Compiled behaviors recover the source-harness gain in 8 of 10 preregistered settings. Where the explicit source pays the heaviest per-call token tax, the compiled form exceeds it. The same bundles transfer zero-shot across tasks (5 of 7 pairs positive) and compose in shift space, while stacked text instructions interfere destructively. Used as a search space, they lift budget-matched harness search by 1.45 times. On three agentic benchmarks the operator aligns policy to the harness con- tract. Under matched data, capacity, and effective strength it outperforms same-capacity LoRA, which needs 11.1 times the parameters to reach the same band, and it regresses nothing where the base policy is already strong. Across 1B-to-8B models the method skeleton is invariant while gain locations move with the model-task pair, which delimits where compiled harnesses should be deployed.

0.40

VLSplat: Vision-Language Guided Object-Centric 3D Gaussian Splatting via Scene Graph

cs.CV cs.AI ·2026-11-01 ·Yuntae Jeon, Younho Jeon, Sujin Jin, Sungho Jo, Seunghee Park
top venuecode/data
Abstract · Accepted to NeurIPS 2026. Project page: https://vlsplat.github.io/

Learning semantic representations for 3D Gaussian Splatting has recently emerged as a promising direction for 3D scene understanding. However, existing 2D-to-3D semantic lifting approaches rely primarily on 2D segmentation supervision and lack object-level priors, causing the learned representation to be driven by local mask evidence rather than holistic object structure. In this paper, we present VLSplat, a vision-language guided approach for object-centric semantic lifting in 3D Gaussian Splatting. Our method augments mask-based lifting with object-level semantic priors derived from vision-language models and refines Gaussian representations using an object-group scene graph. The refinement is performed through two language-guided modules: intra-group refinement suppresses semantic noise within object groups, while inter-group acquisition expands object support regions by incorporating structurally relevant candidates from neighboring groups. Extensive experiments on indoor scene datasets demonstrate that VLSplat improves the semantic consistency and structural completeness of object-level categories. Our code and models are available at https://vlsplat.github.io/.

0.39

Preconditioned Implicit Midpoint Langevin Sampling for Non-Smooth Bayesian Imaging

cs.CV math.NAmath.OC ·2026-11-01 ·Sichen Wang, Zhipeng Lu
top venue
Abstract · Accepted at NeurIPS 2026. 40 pages, 19 figures

We introduce P-IMLA, a preconditioned implicit-midpoint Langevin sampler for high-dimensional log-concave targets with non-smooth regularization. The obstruction is that the large steps enabled by matrix preconditioning are precisely where preconditioned Euler-Maruyama inflates stationary variance. Implicit midpoint removes this surplus via a Cayley-transform identity, making P-IMLA Gaussian-exact at any step size. Beyond Gaussians, whitening transfers the sharp IMLA contraction rate to $κ_{\mathrm{eff}}$, and backward-error analysis gives drift-only rather than Euler-type diffusion bias under explicit smoothness assumptions. An $M$-norm Moreau envelope yields a proximal non-smooth variant with a three-term error budget. Controlled Gaussian and TV-regularized imaging experiments confirm the predicted separation: the unpreconditioned midpoint scheme retains long transients as $κ_f$ grows, whereas P-IMLA remains $κ_{\mathrm{eff}}$-governed and near variance-exact.

0.39

A Bounded Evaluation Protocol for Representational Systematicity

cs.CL ·2026-11-01 ·Boyu Cao, Wenshuo Wang
top venue
Abstract · Accepted at the NeurIPS 2026 Evaluations and Datasets Track

Multiple studies have shown that neural networks can be behaviorally systematic, producing correct outputs for novel combinations of familiar parts. Whether they are also representationally systematic, that is, whether they represent these parts and their roles and use them to produce the outputs, remains hard to test, because each kind of evidence can mislead on its own: correct outputs can come from shortcuts, a decodable signal need not be used, an intervention can act through unintended pathways, and a failed search does not show absence. We propose a bounded evaluation protocol for representational systematicity (Bounded-RS): for an explanation that attributes a systematic behavior to a declared representation, it tests on a matched behavioral contrast whether that representation is recoverable and whether it is causally used, checks the validity of this evidence, and assigns the explanation one of four bounded states, Unsupported, Present-only, Causal-use, or Undetermined, through control-calibrated scores, gates, and stop rules. On a semantic-parsing testbed of 96 matched sets that we release, the tested role-filler Binding explanation of fine-tuned T5 models is Causal-use and the distributed Geometry explanation is Unsupported within its declared search space, verdicts that hold across calibration-compatible parameter configurations, training seeds, and T5-base. On 41 cases with known causal status, full adjudication licenses a false Causal-use in 1 of 34 noncausal cases, against 13 when recoverability is taken as use, without missing any causal case. Bounded-RS thus turns the question of representational systematicity, debated since Fodor and Pylyshyn, into comparable and falsifiable verdicts on declared explanations, each reported with the evidence pattern that licenses it and, for a negative, the search space that bounds it.

0.38

DualClusterNet: Personalized Pattern-Space Composition for Next-POI Recommendation

cs.IR cs.AIcs.LG ·2026-11-01 ·Ziyang Su, Guanjie Zheng, Jilin Hu, Linghe Kong
code/datahot direction
Abstract · submitted to WWW2027

Next point-of-interest (POI) recommendation supports personalized place discovery on location-based Web platforms. Mobility routines can recur across users while being realized at different places and adapted to individual preferences. We present DualClusterNet, a causal sequential recommender that composes two population-shared prototype dictionaries. At each observed prefix, soft routers retrieve pattern and spatial mixtures; a factorized multiplicative interaction couples them, feature-wise affine modulation adapts the shared representation, and a separate user-conditioned branch with an additive--projective gate retains individual variation. The prediction head combines this representation with a learned first-order transition term and ranks the complete training POI vocabulary. We evaluate on NYC and TKY using a training-only vocabulary and an original-final-target protocol that neither bridges unknown POIs nor substitutes earlier targets. The available single-seed results show competitive full-candidate ranking, while controlled analyses support the contribution of the compositional and personalized branches. Code is available at https://anonymous.4open.science/r/DualClusterNet-F4EE/.

0.38

Biological Congruence: Evaluating Microscopy Generation Across Representations

cs.CV ·2026-11-01 ·Duway Nicolas Lesmes-Leon, Shumpei Takezaki, Fabian Schmeisser, Ryoma Bise, Gillian Lovell +3 more
published
Abstract · Accepted at WACV2027

Generative models increasingly produce realistic microscopy images, yet conventional evaluation primarily characterizes visual fidelity and distributional coverage rather than the biological information represented by synthetic data. We introduce biological congruence as an evaluation principle describing whether biologically relevant information forms a coherent description across representations and remains consistent with its empirical biological context. To study this principle, we develop a text-conditioned diffusion framework that jointly generates microscopy images and instance masks, exposing the same generated biological state through conditioning, visual appearance, and instance structure. Across prompt--mask, prompt--image, image--mask, and population-level analyses, we find strongly feature-dependent behavior: density and confluency are consistently preserved, morphology exhibits recurrent disagreement, while nearest-neighbor distance reveals that strong image--mask agreement can coexist with weaker conditioning and population fidelity. Stress testing increasingly unsupported phenotype combinations further shows that target controllability can increase while untargeted phenotypes reorganize and the complete generated state moves toward configurations better supported by real data. These findings demonstrate that generative quality, controllability, cross-representation agreement, and population fidelity capture complementary properties, motivating biological congruence as a relational perspective for evaluating scientific generative models.

0.37

SurgQ-VLM: Question-Conditioned Dual-Stream Compression and Answer-Type-Aware Reinforcement for Surgical Video Question Answering

cs.CV ·2026-11-01 ·Dhruv Makwana, Ananya Mantravadi
publishedcode/data
Abstract · Accepted to WACV 2027 (Applications track). 16 pages: 8-page main paper, references, and supplementary material. Code: https://github.com/Dh…

Reasoning about surgical video -- which phase is underway, which instruments act, and when events occur -- underpins operative documentation, workflow analysis, and skill assessment. Yet surgical visual question answering is still largely built on single frames and closed labels that cannot express a duration, an ordering, or a count, and the video models that lift the frame limit typically compress every clip the same way regardless of the question, discarding the temporal evidence the answer needs. Worse, the answers are structurally heterogeneous, spanning five incompatible structures -- binary, classification, multi-label, counting, and sentence-temporal -- that no single training signal scores uniformly well. We propose SurgQ-VLM, a two-stage model for surgical video question answering. Stage 1 trains a question-conditioned dual-stream (slow/fast) compressor for a LoRA-adapted Qwen2.5-VL-7B: cross-attention conditions both streams on the question, and each clip's visual tokens shrink 49x from the raw ViT patches (11x vs. post-merge backbone tokens). Stage 2 post-trains with GRPO under an answer-type-aware reward that scores each question on the metric matching its answer structure (micro-F1 for label sets, normalised error for counts, a content-timing blend for temporal windows). On a sealed 21,840-question test split over Cholec80 and CholecT50, SurgQ-VLM reaches 0.779 keyword accuracy and a 10x BLEU-4 gain over zero-shot; GRPO adds 2.5 pp K-Acc (2.0 pp attributable to the answer-type-aware reward), concentrated in the structurally hard types, with answer format unchanged (ROUGE-L). Code is available at https://github.com/DhruvMakwana/SurgQ-VLM.

0.37

Spend the Bits Where the Tiny Objects Are: Rate-Utility Region Coding for Aerial Detection

cs.CV ·2026-11-01 ·Yanliang Qi, Jiacheng Ding, Cong Guo, Peihua Mai
Abstract · 17 pages, including supplementary material

Bandwidth-limited aerial cameras must compress their images before a remote detector can find the many tiny objects in them. Threshold-based region-of-interest (ROI) coding sends only confidently detected regions, and because tiny objects provide limited visual evidence and often score low, it can discard the regions that contain them. We propose rate-utility region coding (RUO), which sends a region when its expected detection utility justifies its measured byte cost, choosing to drop it or to send it at low or high quality with one multiplier shared across images. RUO trains no model, uses a standard image codec and decodes in a few milliseconds. On all 1,610 VisDrone-DET test-dev images, RUO uses 42.4% fewer bits than a tuned threshold-based ROI coder, averaged over their shared tiny-object average precision (AP) range. RUO also needs fewer bits for tiny objects than the AV1 Image File Format (AVIF), High Efficiency Video Coding (HEVC) and three published learned codecs, and its advantage over threshold-based ROI coding persists when the sender and receiver use different detectors. For AP over all object sizes RUO needs 25.4% more bits than the tuned ROI coder, because its allocation drops large regions first at low rates, so it suits tasks that target tiny objects. We plan to release the code.

0.37

RCFormer: Reliability-aware Context Transformer for Occlusion-Robust Facial Landmark Detection

cs.CV ·2026-11-01 ·Ashish Kumar, Prashant Kumar, Mallika Garg
published
Abstract · Accepted at the British Machine Vision Conference (BMVC 2026)

Facial landmark detection has advanced substantially under regular imaging conditions, yet occlusion, poor illumination, and large pose changes remain major sources of error because local facial evidence can become incomplete or misleading. Context-guided latent patch recovery offers a promising way to address this issue by inferring unreliable patch representations from surrounding facial context before heatmap prediction. However, the recovered context can itself be unreliable when it is weakly adapted to the input face, influenced by corrupted neighboring patches, or blended without preserving local spatial coherence. We propose RCFormer, a Reliability-aware Context Transformer for occlusion-robust facial landmark detection. RCFormer strengthens latent patch recovery with three reliability-oriented components: visible-context conditioning (VCC), which adapts the context representation using a visibility-weighted global facial summary; visibility-guided context aggregation (VGCA), which reduces the influence of unreliable source patches during attention; and spatial residual correction (SRC), which refines blended latent features using neighboring patch information. The recovered heatmaps are then used as reliability-aware structural cues for existing landmark detection networks. Experiments show that RCFormer improves heatmap recovery and achieves strong landmark localization results on the WFLW, COFW, and 300W datasets.

0.37

MemRole: From Memories to Decisions for Long-Horizon Reasoning in Role-Playing Agents

cs.CL cs.AI ·2026-11-01 ·Mengting Li, Xilong Cheng, Zhihao Xiao, Zhenhua Xu, Jingyuan Zhou +4 more
Abstract · 25 pages

LLM-based role-playing agents are fluent at imitating a character's style, yet struggle to make role-consistent decisions over long narrative horizons. Such decisions are shaped by experiences spanning many prior scenes, which are often partial and may point in different directions. Existing methods inject cross-scene memories as context, but do not reason about how past events shape a character's current motivations and behavior. We propose MemRole, a framework that makes this memory-to-decision process explicit. It introduces Long-Horizon Memory Thinking (LHMT), a reasoning schema that decomposes hidden reasoning into context understanding, memory activation, and response strategy planning, trained on a novel-derived dataset whose character-grounded, timeline-constrained memory pools limit each character to what it could plausibly know at the current narrative point. Beyond supervised fine-tuning, we propose Self-Refining On-Policy Self-Distillation (SR-OPSD): the model critiques its own rollouts against the reference under a data-driven rubric, and a critique-conditioned teacher view of the same model distills token-level guidance back into the policy. MemRole-32B improves over its backbone by 15.2 points on memory-grounded role-playing, performs on par with frontier LLMs such as GPT-5.4, and achieves the best character-motivation recognition accuracy on CroSS-MR.

0.36

Training-Free Motion-Corrected 3D Brain MRI Reconstruction via Event-Guided References

cs.CV ·2026-11-01 ·Wei Jiang, Hongli Chen, Tianyi Ding, Feng Liu, Nan Ye +1 more
code/data
Abstract

Head motion causes artefacts in 3D brain MRI and can bias structural measurements. Model-based correction alternates between image and motion estimation, but artefacts in the reconstructed reference can reinforce motion errors. Learned priors can reduce these artefacts but require pretraining and may add test-time computation. We propose EventMC (event-guided motion correction), a training-free method for interleaved Cartesian acquisitions. It iteratively detects motion events and estimates motion using reference images reconstructed from stable intervals. It then refines the image and motion from low to high spatial frequencies. In a fair comparison in simulation (same reconstruction), EventMC's motion estimates improve mean PSNR by up to 4.0 dB over those of the learning-based MotionTTT. A single run of EventMC is up to 7.5 times faster than MotionTTT with less than half the GPU memory. On in vivo motion-corrupted scans, EventMC achieves similar mean image quality to MotionTTT without any training. Code is available at https://github.com/bayesbreeze/EventMC.

0.36

How Green Is Your Manifesto? Scoring Environmental Positions in the 2024 EU Elections with an LLM Panel

cs.CL cs.CYecon.GN ·2026-11-01 ·Franco Corti, Sajid Siraj
benchmarkcode/data
Abstract

Locating political parties on environmental and climate policy is central to the study of party competition, yet the established instruments, expert surveys and the hand coding of manifestos, are costly, periodic, and slow to extend to new documents. We ask whether a panel of large language models can serve as a cheaper and reproducible alternative. Treating each manifesto as a set of sentences, three models from independent providers score every sentence on two dimensions, environmental protection and climate change, separating a party's position on an issue from the salience it gives it, and the judgments are averaged to the manifesto level. Agreement across the models is retained as a built-in reliability signal. We apply the panel to 41 manifestos from the 2024 European Parliament election, covering Austria, France, Germany, Italy, Spain, and the European party families, and validate it against the 2024 Chapel Hill Expert Survey for the 25 parties with a like-for-like expert benchmark, using the other manifestos as diagnostic cases. The panel recovers the expert ordering of parties with strong rank agreement on both dimensions, while reading them as systematically greener than experts do, a gap we interpret as the distance between manifesto promise and governing record. Inter-model disagreement modestly flags where the instrument errs, and position and salience prove related for environmental protection but independent for climate change. The result is a transparent, low-cost, and multilingual instrument that complements rather than replaces expert and hand-coded approaches; the prompt, code, and manifesto-level results are released openly.

0.35

"PRISM" Is Not All You Need: Cross-Topic Name Collisions and Scientific Paper Retrieval

cs.IR cs.DL ·2026-11-01 ·Jingjie Ning, Xueqi Li, Yibo Kong
Abstract

Scientific names increasingly connect papers from different research topics, leaving readers to recover a paper's identity from context. We study this convergence in 564,861 AI-related arXiv records. Cross-topic name collision grows 4.33-fold from a pooled 2020-2022 reference to 2026, compared with 1.12-fold within topics. A second sentence encoder reproduces the pattern, and name-group resampling quantifies its uncertainty. Fixed word-frequency bands locate the main increase in an intermediate-frequency vocabulary shared by many research areas. We then measure the ambiguity that remains after supplying research clues. With the same three-term query construction, shared-name targets retain matched identification deficits of 25.7 percentage points under title-based BM25 and 6.9 points under title-plus-abstract E5. Within fixed same-name candidates, listwise language-model ranking reaches 89.2% using the query and candidate titles. Topic-specific comparisons show that context resolves more cross-topic ambiguity. The evaluation separates candidate competition, document fields, and contextual ranking through lexical, dense, and language-model baselines. Source-linked cases and a portable evaluation collection make the phenomenon inspectable and reusable for scientific retrieval and research assistants.

0.35

When Does Latent Prediction Error Support Physical Surprise Claims in Video World Models?

cs.CV cs.LG ·2026-11-01 ·Anton Hristov, Aleksandar Slavov, Vasil Vasilev
publishedbenchmarkcommunity pick
Abstract · 28 pages, 12 figures, 25 tables. Accepted at the World Models in Physical AI Workshop, NeurIPS 2026

Prediction errors from video world models are often treated as physical surprise, but they can also reflect appearance, static content, motion, or ordinary prediction difficulty. We present an evaluation framework that separates what frozen representations contain, how they organize appearance and physical differences, and what changes a model's own prediction error. On IntPhys2, a benchmark of physically possible and impossible events, we compare representations from latent prediction, masked pixel reconstruction, and simple visual statistics. Appearance remains easy to decode, while V-JEPA2 exposes modest information about whether events are possible or impossible. Looking at the latent geometry, appearance changes become smaller compared with differences between videos, their directions line up more consistently across scenes, and their overlap with physical changes decreases in deeper layers. Across action-conditioned models, incorrect actions raise prediction error, while appearance changes also produce substantial responses. In Physion++ video clips, repeating a single frame can produce higher error than the original clip, while changing color throughout the clip can lower it. Using Kubric, a simulation and rendering framework, we find that impossible events can produce lower error than possible events. A fixed error threshold also incorrectly flags many possible events as impossible. A model can therefore represent physical information and respond to incorrect actions without its native prediction error reliably measuring physical surprise.

0.35

Cross-Dataset Generalization of Glaucoma Detection CNNs: Effects of Operating Threshold, Multi-Source Training, and Color Normalization

eess.IV cs.CV ·2026-11-01 ·Al-Amin Rahman, Saifuddin Ahmed, Resalat Ahmed
Abstract · 17 pages, 10 figures, 11 tables

Deep learning models for glaucoma detection are usually tested on images from the same dataset used for training. That says little about how they will work at a new hospital or with a new camera. We fine-tuned five ImageNet-pretrained CNNs (ResNet50, DenseNet121, MobileNetV2, EfficientNet-B0, InceptionV3) with identical settings on one of three public fundus datasets (ACRIMA, ORIGA, RIM-ONE DL). Each was tested on a held-out part of its own dataset and on the other two datasets in full. We used three random seeds and report mean and SD over seeds. Mean in-domain AUC was 0.913 and cross-source AUC was 0.702, a drop of 0.211 (SD 0.030; 95% CI 0.135 to 0.286). After correction for multiple comparisons we found no evidence that any architecture did better, though three seeds give low power. At thresholds set on the source validation data, specificity fell from 0.870 to 0.506 and sensitivity from 0.831 to 0.726. Resetting the threshold on 10% of the target images raised specificity to 0.670 but lowered sensitivity to 0.609, and balanced accuracy rose only from 0.616 to 0.639. Training on two sources together did not help (held-out AUC 0.695 versus 0.702). LAB reference normalization (cross-source AUC 0.755) and strong color augmentation (0.754) gave similar gains, cutting the drop to about 0.16. Ben Graham normalization (0.733) and CLAHE (0.712) helped less, and gray-world (0.696) did not help. The two best methods raised specificity but lowered sensitivity. On G1020, used only for testing, AUC stayed near 0.5 in every condition, and networks trained within G1020 were also close to chance. Strong in-domain results do not show that a glaucoma model will work on new data. Normalization and color augmentation reduce the drop but do not remove it. Studies should report external validation and check the operating threshold, not only AUC.

0.35

Advances in Agentic Computer Graphics

cs.CV ·2026-11-01 ·Hengyu Liu, Jihyeon Je, Ruiqi Wang, Mingxuan Wu, Haven Feng +9 more
community pick
Abstract · 27 pages, 19 Figures

Computer graphics is built on code. Large language model (LLM) agents can now generate graphics code and iteratively refine it through various feedback. This capability opens the door to a potentially new paradigm of computer graphics, with graphics algorithms increasingly built around agentic systems to automate progressively complex tasks. Over the past years, research on agentic graphics has expanded rapidly across a diverse range of tasks. This rapid growth makes it increasingly difficult for manual literature reviews to keep pace with new developments. At the same time, similar ideas are often described using different terminology across these tasks, further complicating efforts to develop a systematic understanding of the field. In this report, we address these challenges through two contributions. First, we introduce a unified taxonomy for characterizing agentic systems across diverse computer graphics tasks. Second, we develop an agentic literature system that can assist human researchers in analyzing existing and newly published papers at scale. Building on these contributions, we analyze emerging research trends and discuss open challenges in coordinating agents and tools across graphics tasks, designing interfaces and representations for human-agent collaboration, sustaining long-horizon planning and refinement, and reliably evaluating interactive graphics systems.

0.35

Accurate automated photo-identification of Near Eastern and European fire salamanders

cs.CV q-bio.QM ·2026-11-01 ·Maayan Yesharim, Hava Goldstein, Gabi Kolodny, Assaf Ben-Levy, Tamar Kis-Papo +5 more
Abstract

Photo-identification underpins non-invasive capture-recapture monitoring of amphibians, but manual identification becomes slow and error-prone as reference libraries grow. Focusing on the Near Eastern fire salamander (Salamandra infraimmaculata), we developed a pipeline based on zero-shot local-feature matching (SuperPoint and LightGlue) without species-specific training and evaluated it on $2{,}440$ field photographs from Haifa and Tel Dan in northern Israel. We isolated the yellow spot pattern, scored image similarity by the number of high-confidence keypoint matches, which reduced a confound whereby many-spotted individuals accumulate spurious matches, and assigned identities across the database by two-tier clustering. Pipeline-assisted expert review corrected misidentifications, reducing the number of identities by almost a quarter. On held-out individuals, the pipeline ranked the correct individual first for $95.2\%$ of queries and recovered identities with an ARI of $0.958$. Without modification, it reached $95.1\%$ and $91.7\%$ top-1 accuracy on two datasets of the European fire salamander (Salamandra salamandra) and recognised $93.7\%$ of recaptures in one of them, compared with $38$-$77\%$ for five established programs and $93.9\%$ for visual inspection. Zero-shot local-feature matching thus offers amphibian monitoring a re-identification tool that needs no training data, transfers between related species, scales to databases of over a thousand individuals, and can audit existing databases before capture-recapture analysis.

0.34

Leveraging 2D Foundation Priors for Generalized Category Discovery under Test-Time Adaptation in 3D LiDAR Segmentation

cs.CV cs.LG ·2026-11-01 ·Hanqi Chen, Chaoqi Chen, Tian Ji, Hui Huang
top venuebenchmark
Abstract · Accepted by ACMMM 2026

In autonomous driving, 3D LiDAR semantic segmentation is expected to adapt to evolving environments and recognize previously unseen categories, making generalized category discovery under test-time adaptation (GCD-TTA) an important problem. However, existing works treat GCD and TTA separately, while the limited semantic information in 3D point clouds makes GCD-TTA under domain shift more prone to feature collapse when combining GCD with TTA. To address this issue, we propose a two-stage framework for GCD-TTA in 3D LiDAR semantic segmentation. Specifically, in the first stage, we adaptively aggregate local 3D features according to geometric density variations and obtain reliable pseudo labels under a global class-wise prototype consistency constraint. In the second stage, we cast category discovery as an optimal transport assignment process for globally consistent grouping of novel categories. Meanwhile, semantic features from 2D vision foundation models are fused with 3D point cloud features to enrich 3D semantic information and promote a more separable feature space. Extensive experiments on Synth4D to SemanticKITTI, SynLiDAR to SemanticKITTI, and Synth4D to nuScenes demonstrate the effectiveness of our framework. Under a unified evaluation protocol, our method achieves the best overall mIoU on 9 out of 10 benchmark splits and improves the overall mIoU by up to 12.21 points over the strongest re-implemented baseline.

0.33

Special and General Computational Aesthetics

cs.CV cs.AI ·2026-11-01 ·Ganqing Luo, Chuan Li, Liaoruxing Zhang, Xin Jin
published
Abstract · 18 pages, 5 figures, 2 tables. Presented at the 2026 CCF Chinese Conference on Computational Art (CCAC 2026), Jinan, China, October 2026; to…

Computational aesthetics is a field at the intersection of computer science, artificial intelligence, aesthetics, and other disciplines. As AI has developed, the field has formed relations of inclusion, intersection, or support with computational art, AI art, AI aesthetics, AI-generated content (AIGC), computational creativity, affective computing, and neuroaesthetics. Without distinguishing scope, objectives, and methods, it may be narrowed to image aesthetic quality assessment or expanded into an umbrella term for AI's involvement in art. We analyze its internal structure and its relations with neighboring concepts, and propose a two-level scope: special computational aesthetics (SCA) and general computational aesthetics (GCA). SCA centers on aesthetic measurement and aesthetic process modeling. The former primarily corresponds to "Aesthetic Perception": computationally identifying and representing aesthetic attributes and predicting human aesthetic evaluations; the latter primarily to "Aesthetic Understanding": computationally describing, testing, and explaining how human aesthetic perception and judgment are formed. GCA extends this core to include "Aesthetic Generation," namely artistic generation guided by computational aesthetic models. Some computational aesthetic tasks can also use non-AI methods. Research in this scope using AI algorithms and models we define as AI aesthetics, which may involve the special core and the general extension. "Aesthetic Generation" intersects with AIGC, AI art, and computational creativity, but belongs to the general extension defined here only when computational aesthetic models substantively constrain, guide, or optimize artwork generation. We also draw on Feng Peng's discussion of the computability of beauty, aesthetic experience, and art as a theoretical reference for clarifying boundaries between computational aesthetics and related concepts.

0.33

Spade-SLAM: Sparse Correspondence Anchors for Dense Self-Supervised Visual SLAM

cs.CV ·2026-11-01 ·Julian Philipp Schmitt, Weirong Chen, Ganlin Zhang, Felix Wimbauer, Daniel Cremers
code/data
Abstract

We introduce Spade-SLAM, a self-supervised dense visual SLAM framework that substantially narrows the long-standing performance gap to supervised methods in a zero-shot setting. By demonstrating that high-fidelity tracking can be achieved without explicit ground-truth pose and depth labels, our approach offers a highly effective and label-efficient alternative to fully supervised pretraining. Our core contribution is a complementary sparse-to-dense training regime: While a warping-based photometric loss provides a dense training signal, we stabilize the optimization and suppress drift in ambiguous regions using a sparse correspondence loss derived from image feature matches. Extensive experiments demonstrate that Spade-SLAM significantly improves robustness and accuracy compared to existing self-supervised baselines, closely rivaling SLAM systems trained with explicit pose and depth supervision. The code will be made publicly available at https://github.com/JulianPSchmitt/Spade-SLAM.

0.33

Can AI Reviewers See Your Efforts to Improve Your Paper? A Large-Scale Study of Paper Revision Trajectories

cs.CL ·2026-11-01 ·Yaochen Wang, Chenguang Wang, Ming Li, Haomin Zhuang, Xinyue Zeng +6 more
Abstract · 43 pages including appendices, 8 figures

AI reviewers are increasingly used to assess scientific manuscripts, yet it remains unclear whether their judgments reflect the improvements authors make through revision. We investigate this question using PaperTrajectory-2026, the 2026 release of PaperTrajectory, a dataset of 6,959 paper trajectories across five AI conferences, with automatic annotations of changes along 14 dimensions and separate judgments of overall quality. The annotations show that major improvements emphasize empirical evidence and framing, while later revisions tend toward incremental improvement. Using independent assessments of manuscript versions from eight AI reviewers, we examine whether annotated major improvements receive higher scores, which revision dimensions are associated with score changes, and whether user opinions can outweigh observed revision gains. Every reviewer leaves scores unchanged on most major improvements, and agreement on the direction of score changes is weak. Revision dimensions show some reviewer-specific associations with score increases, but provide limited predictive discrimination and no consistent guide across reviewers. Even when a revision earns a higher neutral score, a brief negative user opinion can erase or reverse that advantage, with substantial variation across reviewers. These findings show that substantive revisions are not consistently reflected in review scores and that observed scoring gains can be vulnerable to user framing, motivating assessments that connect manuscript evidence to overall judgments and remain stable when no scientific evidence changes.

0.33

Peripheral prototypical contrastive loss for supervised medical image segmentation

cs.CV physics.med-ph ·2026-11-01 ·Takuro Shimaya, Ryosaku Shino, Masahiro Saiko
published
Abstract · 5pages, accepted to ISBI2026

Robust segmentation of biomedical images remains challenging primarily due to ill-defined object boundaries. Although several loss functions have recently been proposed to enhance object boundary recognition, concerns remain about the susceptibility to unstable annotations resulting from boundary ambiguities. Here, we introduce the Peripheral Prototypical Contrastive (PPC) loss, which leverages prototypical contrastive learning to improve boundary recognition and enable stable training even under noisy annotation conditions. The PPC loss utilizes prototype features obtained from local regions near object boundaries, focusing specifically on ambiguous areas that have been challenging for existing approaches. Extensive experiments across multiple datasets demonstrated that the PPC loss consistently improve segmentation performance even under conditions with noisy labels. A newly proposed strategy, which adaptively determines the local region for prototype calculation, enhanced robustness and outperformed existing methods. Our findings indicate the potential of the PPC loss in facilitating scalable and reliable medical image segmentation, including the development of foundation models with imperfect annotations.

0.33

Resampling-based batch reweighting for ComBat harmonization of MRI radiomic features in a multi-scanner glioma cohort

eess.IV physics.med-ph ·2026-11-01 ·Rifa Sajda Tasnim, Yushuf Sharker, Md Forhad Hossain, Rakibur Rahman, Enamul Hoque Bhuiyan
published
Abstract · Original article. Submitted to Journal of Magnetic Resonance Imaging (JMRI)

Radiomic features derived from magnetic resonance imaging (MRI) are sensitive to scanners; therefore, they require harmonization across scanners. In a radiomic feature dataset, some batches may be severely underrepresented compared to the others e.g., the Erasmus Glioma Database, which may provide harmonized features biased towards the larger batches in the dataset. We investigated whether reweighting underrepresented scanner batches before empirical-Bayes ComBat estimation improves feature harmonization and reduces residual scanner association while retaining grade-related information. In this retrospective, the Erasmus Glioma Database included 733 eligible patients acquired at a single center across nine vendor-by-field-strength configurations. Structural MRI acquired at 1 T, 1.5 T, and 3 T; four structural MRI sequences were analyzed. From four structural MRI sequences, 428 features were computed using an IBSI-compliant PyRadiomics implementation. The sample sizes of some scanner batches were very small compared to others, which may produce unequal precision and shrinkage of batch-effect estimates. We proposed a data augmentation approach to overcome the limitation of ComBat. We provided exploratory evidence, along with confirmatory Kruskal-Wallis tests with a Benjamini-Hochberg correction to assess the proposed method. Before harmonization, 392/428 features (91.6%) differed significantly across scanner batches (median eta-squared=0.463). Standard ComBat reduced this to 13/428 (3.0%), whereas batch-augmented ComBat reduced it to 8/428 (1.9%); the median residual effect size was <0.001 for both harmonized conditions. Batch augmentation appeared to be a useful strategy to improve the performance of ComBat for feature harmonization under severe batch-size imbalance.

0.32

Spatially-Guided Speaker Extraction Based on Relative Position Cues

cs.SD eess.AS ·2026-11-01 ·Aviad Eisenberg, Shlomo E. Chazan, Sharon Gannot
Abstract

We present a multi-channel target speaker extraction (TSE) framework that is guided by relative spatial cues rather than explicit speaker-dependent references, such as clean enrollment utterances or precise directions of arrival (DOA). The proposed approach encodes the target speaker through discrete, interpretable relations between concurrent speakers, such as left and right or near and far, referred to as the Relative Position Cue (RPC). This relational conditioning effectively leverages spatial information while avoiding explicit DOA estimation, thereby enabling more robust operation under realistic conditions. We evaluate the proposed method across multiple datasets encompassing diverse acoustic conditions, including directional and diffuse babble noise, varying numbers of microphones, mixtures with different numbers of speakers, and dynamic scenarios involving moving speakers. Experimental results demonstrate that the RPC-based approach consistently achieves strong extraction performance and generalizes well across all tested configurations, establishing it as a practical and interpretable alternative for spatially informed speaker extraction.

0.32

EgoMo3R: Joint Egocentric Motion and Scene Reconstruction

cs.CV cs.AI ·2026-11-01 ·Chaitanya Patel, Yusuke Kato, Rin Takahira, Kazuki Kozuka, Juan Carlos Niebles +1 more
publishedcode/data
Abstract · NeurIPS 2026. Project Page: https://chaitanya100100.github.io/EgoMo3R/

Understanding holistic 3D human motion and the surrounding environment from egocentric video is fundamental for applications in AR/VR and robotics. Although these tasks are inherently coupled, existing methods treat them separately. In particular, egocentric motion reconstruction methods typically assume access to metric, gravity-aligned camera trajectories from specialized inertial sensors. Recovering such trajectories from monocular egocentric video remains a significant challenge due to rapid head movements and motion blur. We present EgoMo3R, a joint optimization framework that simultaneously reconstructs 3D human motion and scene geometry from casual egocentric video. It leverages the robustness of low-level visuo-geometric primitives, including monocular depth, optical flow, and point tracks, integrated with a diffusion-based human motion prior via first-order optimization. We propose an alternating optimization scheme that establishes a synergistic feedback loop: scene reconstruction provides physical grounding and trajectory conditioning for the motion prior, while the motion prior regularizes the camera trajectory and anchors the scene in a metric, gravity-aligned world frame. We evaluate EgoMo3R on the highly dynamic sequences of the EgoExo4D and Nymeria datasets and show that it outperforms specialized baselines in both 3D scene reconstruction and egocentric motion estimation.

0.31

Benchmarking Vision-Language Models for Urdu Nastaliq Handwriting Recognition

cs.CV cs.AI ·2026-11-01 ·Haseeb Yaqoob, Fatima Azfar, Emaan Arif Khan
benchmarkcode/data
Abstract

Handwritten Urdu is hard to recognize because the cursive Nastaliq script has no fixed baseline and no clear character boundaries. We benchmark six systems on the Urdu Nastaliq Handwritten Dataset (UNHD): Tesseract (default configuration), a script-specific recognizer trained on printed Urdu (UTRNet-Large), two open-source vision-language models (VLMs), and two commercial VLMs. All six systems are evaluated without fine-tuning on 2,465 single-line images (Part A), and the four VLMs receive the same fixed prompt (decoding uses temperature 0 where configurable, except for Claude Sonnet 5, whose decoding cannot be configured). Both commercial VLMs reach lower error than the printed-Urdu recognizer and the open-source VLMs. We do not compare against handwriting-trained recognizers, which would likely do better in-domain; we characterize zero-shot behavior, not state of the art. We also describe three distinct failure modes among the ranked systems. In one of them, a commercial model explicitly declines to transcribe a few very hard-to-read images instead of guessing. In a second study (Part B), VLMs correct the output of a separate OCR system using both the source image and the noisy text. Once separated from each model's own reading ability, the benefit is small on Tesseract's output (0.7% to 7.4%) and larger (17.9% and 21.9%) only for the two commercial correctors given UTRNet's text. It appears to depend on how legible the supplied text already is, and a weaker corrector can make a more legible text worse.

0.31

Halal-NeSyNet: Clause-Scoped Neuro-Symbolic Cross-Attention for Automated Food Ingredient Auditing and Risk-Aware Halal Logistics Triage

cs.CL cs.AIcs.LG ·2026-11-01 ·Mochamad Rizal Fauzan, Naila Nursyifa, Irgi Surya
benchmarkunusual
Abstract · 14 pages, 7 figures, 13 tables

Verifying that a packaged food is Halal requires reading ingredient declarations and E-number additives before goods enter certified warehouses. Dual-origin additives such as E471, E422, E441 and E920 are Mushbooh (doubtful) unless a botanical, microbial or certified-Halal qualifier appears in the same ingredient clause, so a plant-based qualifier on one clause must not clear an unqualified additive in another. We call a failure to respect this scope cross-clause modifier confounding and build a benchmark of 3,199 items (277 Open Food Facts records and 2,922 template-generated clause-swap items, labelled by an explicit 50-code rule base rather than by a certification authority) with a 540-item counterfactual suite. We propose Halal-NeSyNet, which couples a DistilBERT encoder with a 16-dimensional rule-derived symbolic vector through cross-attention and gated fusion, trained with an asymmetric false-Halal risk loss. Over five seeds, text-only TF-IDF baselines reach 86.25-86.71 % accuracy with 13.42-21.73 % false-Halal leakage and 6.33-26.00 % counterfactual triad consistency, whereas Halal-NeSyNet reaches 97.04+/-0.34 % and 99.67+/-0.50 %. Gradient boosting given the same symbolic vector reaches 96.25+/-1.24 % (+0.59 points from Halal-NeSyNet, 95 % CI [-1.11, +2.28]), so most of the gain comes from clause-scoped symbolic features and not from the network, and the risk loss did not help. Holding out whole template families, text-only models fall to 61.91-78.35 %, Halal-NeSyNet to 90.87+/-11.76 % and logistic regression with the symbolic vector to 91.11+/-2.88 %, with far more spread for Halal-NeSyNet. A three-lane receiving simulation auto-clears 49.33 % of cargo at 0.76 % green-lane contamination. Because labels and features come from one rule engine and 91.3 % of the items are templated, the results measure fidelity to a rule policy, not religious certification.

0.30

FinForesight: A Diagnostic Backtesting Benchmark for Deep Research Systems in Financial Forecasting

cs.CL cs.LG ·2026-11-01 ·Fengyuan Liu, Mukai Li, Jieshun You, Yilin Zheng, Chenjia Min +3 more
benchmarkcode/data
Abstract

Anticipating financial outcomes requires synthesizing evolving evidence and reasoning under uncertainty, making deep research systems promising tools for forecasting. Yet evaluating forecasting ability is challenging because historical backtests may reward memorized outcomes or information retrieved after the forecast date. Live forecasting avoids direct access to realized outcomes, but feedback arrives only after events resolve. To obtain timely feedback and assess historical scores, we introduce FinForesight, a diagnostic backtesting benchmark for financial deep research systems. Across the Chinese, U.S., and Hong Kong markets, the benchmark evaluates four forecasting tasks: valuation, return ranking, event direction, and future financial fundamentals. Two capability diagnostic tasks assess counterfactual reasoning and financial calculation, scored separately from forecasting performance. Our harness-compatible framework restricts retrieved evidence to information available at the forecast date. We compare pretraining-era, resolved post-cutoff, and live evaluations to test whether controlled backtests approximate future performance. Pretraining-era scores exceed post-cutoff scores in most temporal comparisons, and stronger retrieval controls narrow the gap between post-cutoff backtests and live results. Forecasting attribution and robustness analyses reveal how scores respond to retrieval policy and how predictions respond to prompt rewrites and numerical perturbations. The code and benchmark data are publicly available.

0.30

Measuring User Success in Randomized Search Experiments: Interpreting Good Abandonment in Generative AI Search

cs.IR ·2026-11-01 ·Minha Hwang, Shavak Agrawal, David Chen, Carto Wong, Xin Du +3 more
Abstract · 2026 CODE@MIT Conference

Randomized search experiments increasingly evaluate experiences in which large language models answer information needs directly on the results page. Click status alone cannot identify success: a click may lead to an unsatisfactory result, while no click may indicate successful answer consumption, or "good abandonment." Prior research has shown that good abandonment is difficult to detect because dwell, scrolling, and reformulation are meaningful only in context. We present Session Success Rate (SSR), a human-grounded, interpretable outcome metric designed for online controlled experiments. Trained judges label complete session experiences using privacy-protected interaction evidence and video recordings; these offline labels train models scored online from query events, viewport behavior, result-page interactions, and semantic reformulations. Out-of-sample evaluation on 6,904 labeled queries yields 0.75 balanced accuracy overall and 0.68 within abandoned queries; session-level balanced accuracy is 0.68. The sample shows why interaction status is not an adequate outcome: 53.2% of abandonments were successful and 6.5% of clicked queries were unsuccessful. In a validation portfolio of seven ranking and five non-ranking experiments, SSR agreed with Sessions per User in seven cases, was inconclusive in four, and disagreed in one under a common p < 0.05 rule. Separately, across approximately 4,000 production experiments, SSR produced about 2.5 times the binary sensitivity of Sessions per User; in the studied workflow, this supported lower traffic allocations and greater experiment throughput under a fixed traffic budget.

0.30

Q-BWE: Quantum Chaos-Informed Speech Bandwidth Extension

cs.SD ·2026-11-01 ·Tarikul Islam Tamiti, Sajid Fardin Dipto, David Vergano, Luke Baja-Ricketts, Anomadarshi Barua
Abstract · 38 pages, including appendices

High-frequency reconstruction in bandwidth extension (BWE) depends on joined interactions among nonlinear spectral, temporal, and perceptual features; assigning a separate discriminator to each such feature may be redundant and may fail to model their joint interactions. We introduce quantum-informed generative model Q-BWE, where we design a five-qubit Quantum Chaos Discriminator (QCD) instead of using separate discriminator for every features, for modeling nonlinear joined-feature interactions efficiently. Our QCD explores a compact five-qubit quantum-inspired inductive bias based on data re-uploading circuit, trainable controlled NOT (CNOT) rotations, quantum entangling gates, and statevector measurement map. Five features for five qubits are selected from a 26-dimensional speech-dynamics bank and are encoded by a two-layer statevector circuit, producing 31 measurements map of local behavior, correlations, mixedness, and Born entropy. QCD guides a Lattice-ConformerNeXt generator with FiLM conditioning and a cosine-masked residual pathway, while delayed-reward Thompson sampling adaptively controls its contribution. The circuit is simulated on classical hardware. Overall, Q-BWE achieves competitive performance on VCTK English, and MLS French datasets in multilingual and multi speaker setups under clean and noisy conditions across a broad set of evaluation metrics with approximately 1.72x fewer parameters (33.5M vs. 19.4M) compared to recent CIS-BWE baseline published in ACL.

0.30

CNN-Based Retinal OCT Classification: Research Gap Analysis and a From-Scratch Baseline, ED126-A

cs.CV cs.LG ·2026-11-01 ·Shkelqim Sherifi
benchmark
Abstract · 18 pages, 26 figures

Automated classification of retinal pathologies from Optical Coherence Tomography (OCT) images remains challenging because diagnostically relevant morphological patterns may exhibit substantial inter-class similarity. Although high classification accuracies have been reported on benchmark OCT datasets, cross-study comparison is limited by heterogeneous evaluation protocols and by the limited characterization of compact convolutional neural network (CNN) baselines trained from random initialization. This study addresses this gap through two contributions. First, it provides a focused review of CNN-based retinal OCT classification and identifies methodological considerations relevant to controlled benchmarking. Second, it presents ED126-A, a purpose-built four-class CNN baseline trained from scratch under a documented experimental protocol on the Kermany labeled OCT dataset. The model contains 1,241,380 total parameters, including 1,239,460 trainable parameters, and is evaluated using training-dynamics analysis, class-wise performance metrics, confusion-matrix analysis, confidence analysis, t-SNE representation-space visualization, Grad-CAM, hierarchical feature-map analysis, and one-vs-rest ROC/Precision-Recall analysis. On the image-level held-out test set, ED126-A achieved 95.03% accuracy, with a macro-F1 of 0.9205 and a weighted-F1 of 0.9521. The principal class-conditional limitation occurred for Drusen, which attained a recall of 0.9605 but a precision of 0.7493 and an F1-score of 0.8419. The confusion matrix identified Normal to Drusen and CNV to Drusen as major error channels. These results support the use of ED126-A as a documented from-scratch reference baseline and provide an empirical basis for controlled evaluation of transfer learning, increased depth, attention, multi-scale feature fusion, and calibration in subsequent studies.

0.30

MedNet and U$^2$-ViT: A Nested Vision Transformer Diffusion Framework for Diverse Medical Images Generation

cs.CV ·2026-11-01 ·Lakshmikar R. Polamreddy, Jialu Li, Youshan Zhang
code/data
Abstract

The scarcity of diverse, high-quality medical images remains a significant bottleneck in advancing AI-driven medical diagnostics and treatment planning. Traditional datasets often suffer from class imbalances, limited disease coverage, and insufficient variations in imaging conditions - challenges that hinder the development of robust, generalizable models. To address this, we introduce MedNet, a large-scale, multi-modal text-to-image dataset spanning 61 disease categories with 160 classes, covering both humans and animals. We built this with images from trusted public repositories, ensuring data diversity, quality, and ethical compliance. Additionally, we propose U$^2$-ViT, a novel diffusion-based text-to-image generation architecture arranging Vision Transformers (ViTs) in a nested U-Net structure to enhance the generation of diverse, and high-fidelity medical images. Through extensive experiments, we demonstrate that our model achieves the best Fr'echet Inception Distance (FID) score compared to state-of-the-art models. The source code is available at https://github.com/LakshmikarPolamreddy/MedNet-U2-ViT.

0.29

Better Answers, Weaker Evidence Control: Evidence Bypass in On-Policy Distillation

cs.CL cs.AI ·2026-11-01 ·Ruoran Xu, Yilun Zhao, Ziyu Chen, Xiaoqiang kang, Arman Cohan
code/data
Abstract

On-policy distillation (OPD) can improve answer accuracy while weakening responsiveness to supporting evidence. We investigate this separation in scientific-document question answering through matched document interventions that change the supported answer while preserving the question and unrelated context. We identify evidence bypass: a student retains an originally correct answer after the document changes to support a different one. On QASPER with a Qwen3-4B student, standard OPD raises original-document accuracy by 3.4 percentage points, lowers counterfactual accuracy by 5.3 points, and increases bypass by 15.7 points; the same directional pattern appears on PeerQA. Controlled perturbations localize the shift to answer-critical evidence, and bypass grows with the student's pre-distillation closed-book preference for the original answer. A factorized study of teacher targets, privileged information, divergence, and supervision density finds that explicitly supplying the teacher with annotated answer-critical spans yields the largest joint improvement in evidence response and claim support. These findings motivate Evidence-Paired OPD (EP-OPD), which combines on-policy supervision on matched documents supporting different answers to the same question with an answer-level margin that favors each document's supported answer over its paired alternative. Across both datasets and two student sizes, EP-OPD recovers 54-66% of standard OPD's counterfactual-accuracy loss while preserving 80-85% of its original-document accuracy gain. Our results establish evidence responsiveness as a distinct training and evaluation target alongside answer accuracy. Our code is publicly available at: https://github.com/RuoranXu/EP-OPD.

0.29

IncVNM: An Interactive Recommender System with Incremental von Neumann-Morgenstern Utility Learning

cs.IR math.OC ·2026-11-01 ·Jiaxin Wei, Jia Liu
hot direction
Abstract

In modern recommender systems, users are often presented with products whose outcomes are random. The values of such products depend on their entire outcome distributions and on users' attitudes toward risk, making them difficult to evaluate using traditional product-level recommendation methods. We propose \textsc{IncVNM}, an interactive recommender system that incrementally learns a user-specific von Neumann--Morgenstern (VNM) utility function from ordinal click feedback generated during ordinary webpage browsing. At each interaction, \textsc{IncVNM} converts the click order within a displayed product list into pairwise comparisons between outcome distributions and updates the nominal utility by minimizing its deviation from the previous estimate while incorporating the newly observed pairwise comparisons. To close the interaction loop, an entropy-regularized sampling policy balances exploitation of the learned utility against exploration for additional preference information. We evaluate \textsc{IncVNM} in a rolling-window experiment with different response-error probabilities and show that it substantially outperforms the expected-return, mean--variance, discounted-UCB, and distribution-aware BPR baselines.

0.29

Less but Sufficient: Relation Knowledge Selects Compact Knowledge-Graph Contexts, and What External Checks Can and Cannot Catch

cs.CL ·2026-11-01 ·Seunghui Jwa, Minsu Oh, Sangwoo Hong, Yeo Chan Yoon
benchmarkhot direction
Abstract

Large language models (LLMs) read knowledge-graph facts ranked by relevance, leaving open whether they suffice or conflict. Our checks use relation knowledge, written by hand or learned from training questions labelled with their relation paths. It selects a compact context of facts yielding an answer, and derives it; a derivability check tests whether an answer follows, a consistency check whether facts conflict, and a validation-chosen routing gate decides where the derived answer replaces the LLM's. Seven open LLMs read the compact context better in exact match than relevance-ranked facts; on ComplexWebQuestions (CWQ) they match SubgraphRAG's facts at a quarter of the prompt tokens, and stronger LLMs read SubgraphRAG's better. Where the routing gate passes, ours leads SubgraphRAG's in exact match in all 53 LLM, benchmark and prompt settings. A precision gate, returning the derived answer only when the check passes with exactly one derived entity, answered 18 and 5 percent of held-out WebQSP and CWQ questions in a pre-specified test, right for 90 and 96 percent. The consistency check judges facts by path evidence only where another path joins their entities, true of 2.5 percent of named-entity facts in our compact WebQSP and CWQ contexts.

0.29

Detection Instead of Inversion: T1-T2 Microstructure Quantification With Detection Transformers

cs.CV cs.LGphysics.med-ph ·2026-11-01 ·Sebastian Endt, Fatih Özkan, Marion I. Menzel
code/data
Abstract · 4 pages, 3 figures, 1 table; For submission to IEEE ISBI 2027; Code: https://github.com/SebastianEndtTHI/t1t2-detr

T1-T2 correlation MRI has strong potential to quantify microstructural compartments without imposing a biological model. Traditionally, full T1-T2 spectra per voxel are reconstructed with an ill-conditioned and time-consuming Inverse Laplace Transform. We propose to view this challenge as an object detection task: A Detection Transformer (DETR) takes 64 TI/TE signals and directly predicts a set of compartments specified by T1, T2, and signal fraction. We achieve a mean average precision mAP@10 of 76.8% on simulated data, outperforming both our previous model design (53.8%) and regularized NNLS (66.1%). We show that our DETR approach outperforms NNLS in almost every setting across compartment count, compartmental signal fraction, and SNR. Reconstruction is orders of magnitude faster than NNLS. The results are fully reproducible from our public code: github.com/SebastianEndtTHI/t1t2-detr

0.29

Retrieval-Enhanced Cultural Evaluation for Social Media Translation Training

cs.CL ·2026-11-01 ·Linjuan Wu, Yiwen Qiu, Xinze Lyu, Dingming Li, Hui Yao +9 more
code/data
Abstract · 17 pages. Preprint

Social media translation requires preserving meaning while capturing the cultural nuances of user-generated content (UGC). However, gaps in the cultural knowledge of LLM-based evaluators can lead to inaccurate assessments and unreliable training rewards. To improve these signals, we use retrieval-enhanced cultural evaluation to train CUE-MT, a family of social media translation models. A calibrated, domain-specific Judger guides supervised data refinement and is augmented by an Agent for hard-example mining and reinforcement learning. The Agent retrieves relevant external evidence as needed and distills it into concise hints that the Judger uses to score translations and assign rewards. Across 9B, 27B, and 35B-A3B backbones, CUE-MT outperforms the evaluated open-source baselines of comparable size on CUE-MT-Bench and three public benchmarks. Human evaluation on CUE-MT-Bench shows cultural effectiveness gains of 7.44-15.12 percentage points over the strongest external baseline in each size group. Retrieval improves the Judger's four-way classification accuracy by 8.07 percentage points over a matched no-retrieval baseline that uses LLM-generated hints alone. Training ablations further support the combination of hard-example mining and Agent-enhanced rewards.

0.29

Forecasting Earth Surface Change in Embedding Space

cs.CV cs.AI ·2026-11-01 ·Jingtao Li, Sadiq Jaffer, Anil Madhavapeddy, Pedro Sousa, Anthony Ortiz +5 more
benchmarkpositioncode/data
Abstract · 22 pages, Code: https://github.com/ucam-eo/e2forecast

Repeated satellite imagery permits understanding of how our planet changes over time, letting us predict its future and plan mitigating actions. Most existing studies that rely on a time series of satellite observations forecast the spectral values at each pixel for a future target date. However, clouds, shadows, and sensor noise often corrupt these observations, so that spectral forecasters waste much of their capacity on low-level features or noise rather than real change. With the advent of Earth foundation models, we argue that forecasting in embedding space is a feasible and more effective alternative. To this end, we design a token space that handles irregular histories from different sensors and foundation models and supports forecasts at any query date. Tokens from this space are inputs to our proposed forecaster, E$^2$Forecast, which is a transformer with two branches: a Gaussian branch that directly predicts a mean point forecast and per-dimension standard deviations, and a rectified-flow branch that generates forecast ensembles. On Sentinel-1 and Sentinel-2 benchmarks, forecasting embeddings instead of spectral values halves the cosine distance (from 0.20 to 0.10) and greatly reduces overconfidence, lowering the z-score standard deviation from 2.63 to 1.22. By comparing new observations with its forecasts, E$^2$Forecast can also detect anomalous changes without labels or retraining, raising F$_1$ from 41.58 to 48.70 on a global benchmark. This consistent superiority is validated on nine remote sensing foundation models. Beyond change detection, we believe our approach advances the field by permitting effective reuse of diverse pretrained task heads. Our code is publicly available at https://github.com/ucam-eo/e2forecast.

0.29

Learned optical compression for single-pixel quantitative phase imaging

cs.CV physics.opticsphysics.app-ph ·2026-11-01 ·Ary Portes, Yuntian Wang, Che-Yung Shen, Shiqi Chen, Aydogan Ozcan
Abstract · 35 Pages, 11 Figures

Quantitative phase imaging (QPI) reveals optical path-length variations in transparent objects but typically relies on spatially resolved intensity measurements. Here we numerically and experimentally demonstrate learned optical compression for single-pixel QPI, jointly optimizing a diffractive encoder and a digital neural decoder to concentrate reconstruction-relevant phase information into a compact measurement vector. A single reconfigurable phase-modulation plane sequentially displays the learned encoding patterns, which are independent of the input object, each yielding one scalar intensity measurement. Operating at a visible wavelength, the experimental system reconstructs quantitative phase images of unknown objects with ~40-fold dimensional compression, using only 2.55% as many intensity measurements as reconstructed pixels. Experimental blind testing on handwritten digits and grating-like phase objects achieves mean Pearson correlation coefficients of 0.945 and 0.848, respectively. Benchmarking under the same measurement budget demonstrates substantially higher reconstruction fidelity than other encoding strategies, highlighting the advantages of learned optical compression. With the optical encoder fixed, decoder-only neural adaptation enables experimental transfer to previously unseen data distributions, extending the utility of the learned measurement representation. These results establish learned optical compression as a measurement-efficient and reusable framework for QPI, integrating phase-information compression into single-pixel intensity measurements while retaining adaptability to new data. This approach paves the way for low-data-rate phase imaging and sensing, with the potential to extend QPI to spectral regions where high-quality detector arrays are costly or unavailable.

0.29

Learning to read residual images for single-pixel photography

eess.IV cs.AI ·2026-11-01 ·Hongjun An, Shuaizhe An, Tong Tian, Zhe Sun
Abstract · first submit

Continuous video scales in cost with camera count, resolution and recording duration, yet many monitoring tasks seek events rather than complete images and can operate on a compact single-pixel measurement stream. Recovering a chosen interval as video is then desirable, but at 0.24\% of the pixel count the scene correlation image looks like noise and per-clip optimization takes minutes. Correlating the residual of an estimate with the coding patterns nevertheless gives a residual image, the gradient of a centered measurement loss. Lifted through a pretrained video decoder, its agreement with the true error exceeds that of control decoders. Video-latent inversion (VLI) then outperforms frame-wise image-latent reconstruction at 1{,}000 bucket values. Measurement-Gradient-Conditioned Video Flow (MGV-Flow), trained once to read the lifted residual image with preceding frames, reconstructs five frames in 1.55\,s at 22.9\,dB median PSNR, against 398\,s and 18.1\,dB for VLI, with the largest gains in high-motion scenes. Directional fidelity ranks 30 coding configurations and selects one needing a third fewer bucket values. On a physical system, MGV-Flow outperforms per-frame reconstruction without retraining. A sliding window over the bucket-value stream yields 3.2 frames per second on one RTX~4090 and, in theory, 20 frames per second on a data-centre accelerator. Demonstrations are available at https://cavedweller509.github.io/MGV-Flow.

0.29

EarthJev: Calibrated, Non-Generative Answers to Natural-Language Questions about Sentinel-2 Land Cover

cs.CV cs.AI ·2026-11-01 ·Domenico Benfenati, Antonio Maria Rinaldi, Cristian Tommasino
code/data
Abstract · 16 pages, 1 figure, 5 tables. Code and model: https://github.com/CosmoIknosLab/EarthJev

Vision-language models for remote sensing answer questions about satellite images by generating text: every answer costs a decoding loop, has to be parsed back to an option, and comes without a usable probability. We present EarthJev, a 0.8B-parameter model that answers typed questions about multispectral Sentinel-2 patches without generating any token. All 12 bands enter the vision encoder of Qwen3.5-0.8B through a widened input layer initialised from the RGB kernels, and, following the open OneJev recipe, the answer is read from the logits of the option letters at a single position and trained with cross-entropy plus Brier score. Under a pre-specified analysis plan, we compare it with the same backbone fine-tuned to generate the answer and with three multimodal language models fine-tuned with LoRA, all trained on the same 70,890 questions from 5% of BigEarthNet v2.0 and timed in one job on the same GPU; VQA models that classify over a fixed answer vocabulary are reported as a reference for the cost of answering only the questions seen in training. EarthJev answers a question in 19.9 ms, 4.3 to 8.7 times faster than the multimodal language models and 6.8 times faster than its own backbone used generatively; when all the questions about an image are asked together, it answers them in 42.8 ms, 3.7 to 10.1 times faster than the batched generative models. On 36,028 test questions it reaches 86.0% accuracy (three seeds), 0.6 and 1.2 points below the two 2B multimodal models and 3.5 points below the remote-sensing EarthDial-4B, with answers that are valid by construction and calibrated probabilities (ECE 0.006). Replacing the nine non-RGB bands with their mean costs 4.5 points, so the model uses the multispectral input, and in an ablation training the whole vision encoder raises accuracy to 88.4% at the same cost. Code and model: https://github.com/CosmoIknosLab/EarthJev

0.28

Lie Routing Vision Language Models for View-Robust 3D Understanding

cs.CV cs.LG ·2026-11-01 ·Jingjun Yi, Chen Hu, Qi Bi, Hao Zheng, Huimin Huang +5 more
top venue
Abstract · NeurIPS 2026 Accepted Paper

Multimodal Large Language Models (MLLMs) have made substantial progress in 3D scene understanding from multi-view RGB observations. Yet a fixed object and query can elicit inconsistent target-related responses across camera viewpoints, a phenomenon we characterize as viewpoint-induced semantic drift. We introduce LieVLM, a Lie routing framework for view-robust 3D understanding. Motivated by the rotational component of viewpoint change, Lie routing adapts visual representations by applying local $\mathrm{SO}(3)$ transformations to projected features within learned three-dimensional subspaces. Geometry tokens extracted by VGGT predict local rotation parameters and channel-wise routing coefficients. A closed-form Padé--Cayley map transforms projected visual features, whose identity-relative corrections are projected back and gated into the original visual and geometry features. This design retains the pretrained semantic and direct spatial context while learning view-dependent feature adjustments. Experiments on 3D captioning, grounding, detection, and spatial reasoning show consistent improvements over additive fusion across training scales and backbones. Controlled ablations examine the roles of geometric conditioning and transformation structure. Viewpoint analyses further reveal systematic variation in the predicted routing states alongside more localized target responses and stronger task performance across viewpoint changes.

0.28

Can Large Language Models Extend Their Agentic Capabilities to Goal Pursuit in Open-Ended Multi-Turn Interaction?

cs.CL ·2026-11-01 ·Mingyu Chen, Jingkai Lin, Xiaofen Xing, Liankai Huang, Ruiqi Xiong +1 more
Abstract · 18 pages, 9 figures, 20 tables, including appendices

Goal pursuit in open-ended dialogue requires agents to adapt their actions to interaction feedback across successive exchanges. We investigate whether agentic capabilities developed for multi-step tool use carry over to dialogue organized through the model's native tool interface, and how this changes interaction behavior. We define Tool-Mediated Interaction (TMI) as a training-free protocol that represents dialogue actions as tool calls and counterpart responses as tool results, maintaining a continuous reasoning-action-observation loop with retained reasoning and interaction history. We evaluate TMI across social interaction, emotional support, persuasion, and price negotiation using two models and 15,678 main-experiment episodes. Across both models, TMI improves social goal attainment, emotional-support preference, donation amounts, and buyer negotiation margins over conversational baselines with and without thinking. A Qwen ablation further shows that TMI outperforms ordinary conversation with retained reasoning on all four primary outcomes. Behavioral analyses associate the improvements with task-specific patterns in initial proposals and responses to interaction feedback. Alongside these gains, TMI reduces mean reasoning tokens per decision by 42.0-81.5% relative to the thinking baseline and lowers mean cross-turn phrase reuse across all eight model-environment pairs. These findings show that interaction organization shapes both goal attainment and reasoning across dialogue turns.

0.28

Event-Based Optical Flow Constraint via Time-Sliced Plane Fitting

cs.CV eess.IV ·2026-11-01 ·Masaki Koike, Yusuke Kameda, Shunichi Sato, Takayuki Hamamoto
published
Abstract · 5 pages, 5 figures, 1 table. Accepted for publication at IEEE VCIP 2026 (2026 IEEE International Conference on Visual Communications and Ima…

Event-based optical flow estimation using spatiotemporal planar approximation inherently assumes constant velocity and tends to average out temporal velocity variations. In this paper, we propose a flow estimation method that extends this planar approximation along the temporal axis. The resulting event-based optical flow constraint allows existing variational optical flow formulations to model temporal motion variation. Specifically, we partition the space-time volume into sub-intervals and apply a two-step planar approximation to track object edges consistently. Moreover, a spatial compensation mechanism and a temporal regularization term are incorporated to compensate for edge displacements and temporally stabilize the flow estimation, respectively. The experimental results show that modeling temporal motion variation improves flow estimation accuracy over conventional methods.

0.28

UA-Depth: Uncertainty-Aware Depth Restoration for Material-Agnostic Object Manipulation

cs.CV cs.RO ·2026-11-01 ·Rui Cao, Jia-Hui Pan, Yin Zhang, Qiwei Meng, Jiapeng Li +6 more
benchmarkdataset
Abstract · Under Review

Robotic manipulation requires accurate metric depth, yet commodity RGB-D cameras often fail on non-Lambertian objects such as transparent, reflective, and translucent materials, producing holes, structured noise, and biased depth. Most restoration methods produce only a single depth map, treating noisy and reliable regions uniformly for downstream tasks while ignoring regional uncertainty. We present UA-Depth, an uncertainty-aware depth restoration framework that iteratively co-refines depth and pixel-wise uncertainty, and uses uncertainty-gated features to suppress unreliable cues for stable updates. To strengthen zero-shot behavior, UA-Depth leverages a frozen depth foundation model as a prior to initialize missing or unreliable depth, and applies two-stage MAD-weighted log-depth scale normalization to reduce outlier bias in relative-to-metric alignment. We introduce an eight-dataset zero-shot benchmark across diverse scenes, materials, and sensors. Across these benchmarks, UA-Depth achieves the best mean rank among representative baselines, with particularly strong gains on datasets dominated by transparent and reflective objects. Beyond depth metrics, we demonstrate downstream improvements in category-level 6D pose estimation and 6D grasp synthesis by exploiting the predicted uncertainty. \emph{Code, models, and data for training and evaluation will be released upon publication.}

0.28

Multi-Agent Collusion and the J-Space

cs.CL ·2026-11-01 ·Marco Molinari, Gabriele Bettineschi
benchmarkhot direction
Abstract

LLM agents are increasingly deployed in Multi-Agent Systems (MAS), and they introduce the risk of coordination among agents that might escape human oversight. The J-space has been used to uncover deception in single LLMs, but its use for uncovering and steering collusion between agents remains unexplored. We show, with interpretability experiments on DeepSeek-V4-Flash (284B) on the NARCBench collusion benchmark, that colluding agents' J-space holds the collusion itself: before replying, 86% of colluders represent misrepresentation, concealment or remorse. In a steganographic card-counting task, the receiver's J-space holds a hidden count, and causal interventions alter its count. We also introduce the Jacobian Pair Contrast (J-PACT), an unsupervised method for detecting collusion between agents. We evaluate J-PACT on four open-weight models: GPT-OSS-20B, Llama-3.3-70B-IT, Qwen3-32B, and DeepSeek-V4-Flash (284B). J-PACT retains 85% of the above-chance AUROC of supervised NARCBench probes, matches them on steganographic scenarios, and names the colluding pair up to its complement in 99% of detected runs.

0.28

MMCoT$^3$: Empowering 3D Scene Understanding with 3D-Text Interleaved Multi-Modal Chain-of-Thought Reasoning

cs.CV ·2026-11-01 ·Yongqi Shan, Yifan Wang, Yunzhi Zhuge, Lijun Wang, Huchuan Lu
top venue
Abstract · 10 pages, 5 figures, 4 tables. Accepted to ACM Multimedia 2026

Understanding and reasoning over 3D scenes from point clouds is essential for a broad range of AI applications. Recent 3D large language models have shown promising results, yet they still face two fundamental limitations: (i) insufficient capability for grounded multi-step reasoning in complex 3D scenes, and (ii) a text-centric inference paradigm that lacks effective interaction between textual reasoning and visual evidence. As a result, these models often fail to verify intermediate inferences, disambiguate similar object instances, and avoid hallucinations. To address these limitations, we propose MMCoT$^3$, the first 3D-Text Interleaved Chain-of-Thought (CoT) framework for 3D scene understanding. The core idea is to explicitly interleave textual reasoning with task-relevant 3D evidence, enabling the model to consult scene observations throughout the reasoning process. In addition, since a textual mention in 3D scenes may correspond to multiple candidate objects, naive object-mention alignment can inject incorrect evidence and mislead subsequent reasoning. To mitigate this multi-instance ambiguity, we design an ambiguity-aware evidence aggregation mechanism that uses query-guided cross-attention to fuse candidate object tokens and inject both evidence and ambiguity cues into the reasoning process. To better equip the model with multi-step reasoning capability, we further develop PointCoT, a large-scale 3D-specific CoT instruction tuning dataset with approximately 176.2K instances spanning captioning, question answering, dialogue, and planning tasks. Extensive experiments on six benchmarks show that our method consistently improves performance and achieves state-of-the-art results, validating the effectiveness of our proposed ambiguity-aware 3D-text interleaved reasoning for 3D scene understanding.

0.28

EEG-HealthLLM: Hierarchical Adaptation of Large Language Models for Unified Heterogeneous EEG Health Assessment

cs.CL ·2026-11-01 ·Ruidong Ma, Jisheng Dang, Ronghao Lin, Bimei Wang, Jingwen Zhao +5 more
code/data
Abstract · 15 pages, 7 figures

Electroencephalography (EEG) provides a non-invasive means for neurological and mental health assessment, yet existing EEG models are typically developed for individual tasks and are difficult to unify across heterogeneous signal configurations and assessment objectives. Large language models (LLMs) offer a common interface for multiple tasks, but adapting heterogeneous EEG signals to a shared LLM raises a key challenge: how to preserve task-specific signal characteristics while enabling effective language-side parameter sharing. To address this problem, we propose EEG-HealthLLM, a unified and parameter-efficient framework for heterogeneous EEG health assessment. We introduce an Asymmetric Hierarchical Adaptation Strategy: task-specific EEG adapters preserve heterogeneous spatio-temporal dynamics, while a unified shared LoRA is injected into the LLM backbone (e.g., Phi-3) to capture shared high-level assessment semantics across tasks. We further establish EEG-HealthBench by reorganizing five existing EEG resources into paradigms covering epileptic seizure detection, major depressive disorder assessment, anxiety risk assessment, depressive symptom risk assessment, and emotional state assessment under a unified subject-independent protocol. Extensive experiments show that EEG-HealthLLM achieves an average balanced accuracy (BAC) of 69.58%, ranking first among the evaluated methods in both average BAC and macro-F1. Compared with independent single-task LoRA adaptation, it improves average BAC from 67.70% to 69.58% while reducing the aggregate language-side adapter footprint by 80.0%. Further analyses reveal task-dependent cross-task interactions and trade-offs, supporting the proposed hierarchical adaptation strategy. The implementation is publicly available at https://github.com/Marui-88/EEG-HealthLLM.

0.28

SVD-Prune: Spectral Vision Token Pruning under Extreme Compression

cs.CV ·2026-11-01 ·Yvon Apedo, Martyna Poreba, Michal Szczepanski, Samia Bouchafa
publishedpositioncode/data
Abstract · Accepted at WACV 2027. Code available at https://github.com/anoncvlab/svdPrune

Vision-Language Models (VLMs) and Vision-Language-Action (VLA) models process hundreds of visual tokens, making visual representations a major computational bottleneck. We argue that visual token pruning should be formulated not as identifying individually important tokens, but as preserving the dominant low-rank spectral structure of the visual representation. Based on this formulation, we introduce SVD-Prune, a training-free framework that leverages Singular Value Decomposition (SVD) and statistical leverage scores for globally informed token selection. SVD-Prune is architecture-agnostic and can be applied at different stages of the multimodal pipeline, including encoder-side, decoder-side, and hybrid configurations Extensive experiments across VLM and VLA benchmarks demonstrate strong robustness under extreme compression. SVD-Prune reduces the visual token. SVD-Prune cuts the visual token budget by up to 99.3%, while, at a 16-token budget, encoder-side pruning lowers inference FLOPs by 82.4% and achieves a 4.18x end-to-end speedup, including pruning overhead.

0.28

Z-Cache: Accelerating Diffusion Transformers via Self-Reflection

cs.CV ·2026-11-01 ·Zegang Cheng, Zhikai Wang, Jiacheng Liu, Xiaobing Tu, Chang Zou +6 more
publishedcode/data
Abstract · Accepted at NeurIPS 2026 (poster)

Diffusion transformers have become the most powerful models for visual generation, but still suffer from massive computation costs. To solve this problem, feature caching has been proposed to cache the features of diffusion models in the previous computation steps and then reuse them in the following caching steps, which brings significant acceleration but also degradation in generation quality. To address this problem, this paper proposes Z-cache as a feature caching method that can maintain high-quality generation through self-reflection. Concretely, we observe that the error from feature caching tends to be sharply reduced after each full computation. Based on this observation, Z-Cache is designed to first predict the features in the future caching steps and then perform a full computation. After that, Z-Cache returns to the caching steps and re-predicts them based on the previous and the current computation steps, which brings correction in features. Experiments demonstrate that with Z-Cache, diffusion transformers achieve comparable generation quality to the original model but with faster inference speed, for instance, 6.22X acceleration on FLUX-dev for text-to-image generation. Our code is released on GitHub.

0.28

UniGS-SLAM: Monocular Gaussian SLAM with Enhanced Geometry and Open-Vocabulary Semantics

cs.CV ·2026-11-01 ·Xingyuan Ding, Sicheng Yu, Kaiyong Zhao, Hao Wang
code/data
Abstract · 23 pages, 9 figures

Monocular semantic SLAM aims to reconstruct scene geometry and build semantic maps from RGB observations. Existing Gaussian-based methods often attach semantic labels or features to evolving Gaussian primitives, making semantic fusion difficult to maintain and increasing reconstruction overhead. Hybrid Gaussian--voxel representations provide more stable semantic support, but often rely on RGB-D input or known poses and do not revise historical semantic evidence after geometry updates. We present UniGS-SLAM, an RGB-only framework that decouples open-vocabulary semantics from Gaussian reconstruction while preserving bidirectional interaction. Foundation priors enhance monocular depth and pose estimation, while semantic observations are fused into an independent sparse voxel map with observation provenance, enabling historical contributions to be revised as reconstruction evolves. The fused semantics further guide Gaussian filtering and local surface refinement. Experiments on Replica, ScanNet, and KITTI-360 show leading tracking, rendering, and geometric reconstruction performance among representative RGB-only methods, together with strong open-set semantic mapping on Replica and ScanNet. We also achieve a 2.37-fold reconstruction speedup over the fastest compared semantic SLAM baseline. Project page: https://3dagentworld.github.io/UniGS-SLAM/.

0.28

Band2Net: Frequency-to-Network Transport Distillation for fMRI-Privileged, EEG-Only Session-Condition Decoding

cs.CV ·2026-11-01 ·Nishi Kanta Paul, Md Shihabul Islam Shovo, Dwarikanath Mahapatra
Abstract · 5 pages, 2 figures, 1 table

Functional MRI resolves large-scale brain networks but is expensive and stationary, whereas EEG is portable but poorly localised. We treat fMRI as privileged information, available during training and absent at deployment, with EEG and fMRI acquired from the same participants under matched conditions but not simultaneously. Band2Net encodes EEG into five frequency-specific connectivity graphs and fMRI into seven canonical network tokens from a frozen teacher, and solves a balanced entropic transport problem between the two token sets that gives each EEG band its own fMRI target. In subject-grouped cross-validation on the PsiConnect psilocybin data set, Band2Net reaches 0.801 ROC-AUC for EEG-only session-condition decoding, 5.0 AUC points (95% CI 1.8-8.2) above an EEG-only student. Although its margins over a single shared plan, a KL-relaxed plan and label-free network descriptors remain unresolved, secondary comparisons favour band-specific transported targets over logit, global-feature, contrastive and pooled-target distillation, though not after Holm correction.

0.27

Learning from high-resolution 2D images to reconstruct 3D microstructures

cs.CV ·2026-11-01 ·Wenbo Sui, Lars P. Mikkelsen, Jens W. Andreasen, Daniel Lichau, Harold Phelippeau +1 more
unusual
Abstract

High-resolution three-dimensional (3D) microstructure characterization remains difficult to undertake at scale because volumetric imaging at the required spatial resolution is expensive, time-consuming or destructive. In many settings, however, fine-resolution two-dimensional (2D) imaging and lower-resolution volumetric measurements are both readily available. We introduce a Privileged Cross-modal Relational Distillation (PCRD) framework that leverages fine-resolution 2D observations during training to improve specimen-specific 3D reconstruction from lower-resolution volumetric imaging alone. We demonstrate PCRD on pultruded carbon fiber-reinforced polymer composites, using scanning electron microscopy as the privileged modality and micro-CT as the inference input. The framework transfers spatial organization by matching similarity distributions defined around corresponding spatial anchors in the teacher and student representations, avoiding direct alignment of heterogeneous feature vectors. In 2D cross-sectional reconstruction, PCRD improves mAP by 1.66 percentage points over direct micro-CT training. Independent 3D validation against high-resolution synchrotron X-ray CT data shows close agreement in fiber volume fraction, diameter and local spacing. The PCRD reconstruction matches 89.97% of the reference fiber centerline length, compared with 75.08% for the interpolated low-resolution volume.

0.27

Understanding the relationship between retrospective soundscape attributes and continuous pleasantness and eventfulness ratings

eess.AS cs.SD ·2026-11-01 ·Yuanbo Hou, Kening Guo, Francesco Aletta, Woon-Seng Gan, Dick Botteldooren +1 more
unusual
Abstract

Soundscape studies ask listeners to rate perceived affective quality (PAQ) attributes, such as pleasant and vibrant, after listening to recordings or experiencing acoustic environments. These retrospective ratings summarise an experience during which impressions may change. International Organization for Standardization (ISO) procedures convert these PAQ ratings into Pleasantness and Eventfulness scores. Pleasantness and Eventfulness (P/E) can also be rated directly and continuously during listening, capturing changes in reported perception over time. Retrospective PAQ ratings and continuous direct P/E describe the same soundscape experience from different perspectives. We examine how PAQs relate to direct P/E and how ISO-derived P/E corresponds to direct judgements, using 1,782 paired 30-s trials involving audiovisual recordings from 18 streets. The results show that PAQs retain a predominantly two-dimensional structure. Vibrant and Monotonous lie closer to the Eventfulness-Uneventfulness axis in the direct P/E plane than in the ISO 2025 reference geometry. The full 30-s direct P/E mean has lower calibrated error against ISO-derived P/E than three 10-s means. An empirical mapping using the same 8 PAQs reduces held-out joint root-mean-square error by 4.9% relative to affine-calibrated ISO 2025 scoring. The findings show that retrospective and continuous assessments offer related but distinct soundscape descriptions.

0.27

FlyWheel: Attention Kernels Built for the TPU

cs.CL cs.LG ·2026-11-01 ·Zishuo Bao, Zeshen Zhang, Linkun Chen, Yucheng Lu
Abstract · 24 pages, 10 figures, 7 tables

Attention kernels on Tensor Processing Units (TPUs) often underutilize their matrix units (MXUs) because of repeated matrix operand loads, vector-intensive computation, and dependencies that leave MXUs idle. We present FlyWheel, a set of fast TPU kernels for both softmax and linear attention. FlyWheel reformulates softmax and triangular inversion to reduce vector work, while coordinating data layouts to improve matrix operand reuse. It then overlaps independent computation to hide execution latency. We apply these techniques to both softmax attention and Gated DeltaNet on TPU v6e and v7x. FlyWheel speeds up softmax attention by up to $4.14\times$ over the state-of-the-art RPA kernel and $1.80\times$ over Splash Attention, reaching 81% of peak BF16 throughput on v6e. It also speeds up Gated DeltaNet prefill by up to $3.61\times$. Integrated into vLLM serving Qwen3.8-27B, FlyWheel improves prefill throughput by up to $1.56\times$ while maintaining comparable accuracy.

0.26

Scene-Adaptive Readout and Decoding in Diffusion Driving VLAs

cs.CV cs.AIcs.RO ·2026-11-01 ·Seungjun Yu, Jiho Choi, Inha Kang, Hojun Choi, Seungwook Kim +1 more
code/data
Abstract · The manuscript requires substantial revisions to its methodology, experimental analysis, and presentation of the findings. The authors belie…

Diffusion driving vision-language-action models repeatedly execute the backbone at every denoising stage, incurring substantial inference cost, while it remains unclear whether all scenes require the same decoding computation. To distinguish when a trajectory becomes readable from how much further refinement it requires, we analyze trajectory readout across network depth and decoding progress. We identify a depth-decoding dissociation in which late-layer representations already support useful trajectory readout after the first forward sweep, whereas early-layer representations benefit more from continued decoding. We further find that both the preferred readout layer and decoding depth vary across scenes. Based on these observations, we propose a scene-adaptive readout and decoding framework. A scene-conditioned readout MoE adaptively weights and combines trajectory predictions from layer-specific late-layer representations after a single backbone sweep, while an adaptive decoding policy applies additional native refinement only when required by the scene. We evaluate our framework on two diffusion driving VLAs across the Waymo E2E and nuScenes benchmarks. Our method improves validation RFS by up to 0.56 while also improving test performance, and achieves an end-to-end speedup of up to $6.0\times$ on Waymo E2E, while consistently reducing trajectory error on nuScenes. These results demonstrate that scene-level readout and decoding allocation improves planning quality while avoiding unnecessary decoding. Our code is available at https://github.com/sjyu001/SARD

0.26

Systematic Evaluation of Robustness in Fine-tuned Models

cs.CV cs.AI ·2026-11-01 ·Angela John, Hamayoon Behmanush
survey
Abstract

Our study systematically evaluates the robustness of fine-tuned pre-trained models across different datasets and attack types, aiming to provide insights for researchers in selecting pre-trained models for image classification tasks while prioritizing robustness against potential attacks. Our work explores the influence of fine-tuning pre-trained models with different datasets and attack types on their robustness and overview of the appropriate pre-trained model for image classification based on factors such as computation resources, memory usage, and accuracy on clean data. We fine-tune the models with weights. Evaluate the models on data before any attack and after different attack cases. The study utilizes publicly available image datasets such as MNIST, CIFAR10, CIFAR100, and SVHN to assess the robustness of pre-trained models. The results indicate that the robustness of fine-tuned models varies across different data distributions and sample sizes, emphasizing the need for tailored selection based on specific task requirements. The findings also highlight the varying performance of ResNet50, DenseNet, and MobileNet in defending against attacks on different datasets. Overall, this research provides valuable insights into the robustness of fine-tuned pre-trained models and their suitability for real-world applications in image classification.

0.26

CCD: Context-Calibrated Open-Entity Out-of-Distribution Detection in In-Context Tabular Foundation Models

cs.IR ·2026-11-01 ·Zhangkai Wu, Xu Zhang, Zhihao Hao, Zhihong Cui, Yongmin Yoo +2 more
Abstract

In-context tabular foundation models (TFMs), such as TabPFN, predict from a labelled context without updating their parameters and are increasingly used for out-of-distribution (OOD) detection in Web applications, whose test rows come from entities absent from that context. Such applications need open-entity OOD detection: a new entity must not be mistaken for an unknown class, and an unknown class must still be caught. Existing detectors fail in this setting, and an entity-split gap hides the failure: random splits also place the test entities in the context. The failure arises because every TFM score reflects how unfamiliar a row is to the context. This produces an entity-over-class bias: known-class rows of new entities look more OOD than familiar ones, sometimes more than unknown classes. Specifically, the two OOD scores most widely used with in-context TFMs face an entity-class trade-off: the TFM's predictive confidence misses unknown classes, whereas the distance of a test row to the context flags new entities. To address this problem, we hold a group of entities out of the context and score them with the remaining context, which turns them into new entities of known classes and tells us how such rows should score. To measure detection on new entities, we build NEO-Tab (New-Entity OOD detection on Tables), 14 grouped and temporal tables on which all test rows come from new entities. We then propose Context-Calibrated Detection (CCD), which uses these group-out swaps on in-distribution data to calibrate three distances and the confidence against new-entity references. CCD weights them by their sensitivity and sharpness, keeps the model frozen and uses no OOD data. On NEO-Tab, CCD outperforms 16 existing detectors on every metric we report, and its lead holds on three TFM backbones.

0.26

Task-driven sparse-view planar X-ray computed tomography for axial metrology

eess.IV cs.CV ·2026-11-01 ·Yibin Tian, Shuming Yang, Wolfgang Heidrich
Abstract

Computational imaging can reduce the acquisition burden when the system is designed around task-relevant information rather than complete image recovery. We investigate this framework for sparse-view planar X-ray tomography (XCT) of layered objects, using measurement of the axial termination gap (ATG) of back-drilled printed-circuit-board (PCB) vias as a representative task. Instead of using 3D reconstruction, the proposed method estimates ATG directly from polychromatic projection data through a physics-based sensitivity model. A gated hierarchical estimator first fits ATG together with radiometric perturbations and selectively introduces global geometry and specimen-warp models only when a reduced chi-squared adequacy test and Bayesian information criterion justify the added degrees of freedom. The same low-dimensional perturbation model is incorporated into marginalized Fisher-information analyses, yielding task-specific conditional and calibration-aware Cram'er-Rao bounds. Numerical experiments show that, under a fixed total photon budget, task-driven metrology changes little over 12-64 views, while reconstruction-driven metrology is strongly view-dependent with much larger errors. The direct estimator also follows the expected photon-limited scaling. These results demonstrate a computational sensing strategy in which XCT acquisition, physical modeling, and inference are evaluated according to the quantitative information required by the task rather than for general-purpose volumetric image fidelity.

0.25

The Structure of Disagreement in Emotion Annotation

cs.CL ·2026-11-01 ·Emily Öhman, Anna Koufakou, Eugenia Diegoli
position
Abstract

Calculating inter-annotator agreement is an established practice in NLP, and disagreement-aware work increasingly retains the distribution of human judgments rather than reducing annotations to a single gold label. Keeping the distribution, however, does not explain why annotators disagree. This paper examines the types of disagreement that occur in emotion annotation, starting from multilabel annotations of 1,677 sentence-level units from Ernest Hemingway's The Old Man and the Sea and testing the patterns they reveal against individual-level annotations from BRIGHTER and GoEmotions. Across the three datasets, we find three main recurring disagreement structures. Emotion categories overlap systematically; annotators differ both in how broadly they apply labels and in how they treat particular category boundaries, while sparse participation and explicit abstention can look like disagreement. Where intensity is rated, disagreement over whether an emotion is present can be separated from disagreement over its intensity once recognized. Similar aggregate disagreement can therefore arise from qualitatively different annotation processes. Hard aggregation removes most of this structure, while annotator vote-share representations keep marginal label frequencies but not the co-occurrence, annotator-specific, or procedural patterns that produced them. We argue that disagreement should be understood in terms of its underlying patterns before it is summarized, aggregated, or used as a modeling target, and propose a vocabulary for disagreement structure.

0.25

Shared relations, specific choices: conditional sensitivity and competing pathways in language models

cs.CL ·2026-11-01 ·Kaixiang Shu
Abstract

How does a relational representation shared across queries drive a query-specific choice? Controlled entity-binding experiments with GPT-2 Small, Qwen2.5-1.5B and Llama-3.2-1B reveal a low-dimensional relational background that is predominantly shared across query roles and transfers to new names. Same-domain interactions persist when another attribute domain is queried. After removing this component, restoring independently trained relation directions brings output contrasts closer to their original values than sign-reversed or equal-norm random controls in all three models. Yet a transferable relational representation need not have a transferable functional sign. In Qwen2.5, fixing model weights, the training-mean relation vector and edit dose yields opposite selection effects across binary and lexical-mapping prompts with different answer vocabularies. A factorial contrast of downstream sensitivities across attribute-query combinations determines the shared direction's local signed selection gain. In a separate eight-condition protocol, mean local gain is positive in Qwen2.5 and negative in Llama3.2. Late-component clamps resolve opposing pathway contributions: in Llama3.2, negative transmission through selected attention-only paths is partially offset by positive transmission through paths traversing both selected attention and MLP sets. Source-coordinate analyses establish distributed factor coverage, hidden-receiver readouts reveal layout-dependent combination, and instance-geometry interventions separate response direction from amplitude. These findings show how conditional sensitivity and competing pathways give shared relational structure its effect on specific choices.

0.25

Text-Assisted Event Stream Visual Place Recognition via Multi-Scale Spatiotemporal Optimal Transport

cs.CV ·2026-11-01 ·Lan Chen, Xingxing Xiong, Xiao Wang, Ziyan Zhang, Zikang Yan +3 more
Abstract · 17 pages including appendix, 4 figures. Submitted to ICLR 2027

Benefiting from the high dynamic range, high temporal resolution, and low energy consumption of event cameras, low-altitude economic research leveraging event cameras has become a popular topic. Several existing attempts introduce natural-language descriptions into visual place recognition (VPR) to mitigate the sparsity of structural spatial information inherent to event cameras. Nevertheless, these methods still fail to fully address two core challenges: effectively encoding event streams and semantically aligning event streams with text. In this paper, we propose a multi-scale spatial-temporal optimal transport approach, MSTOT-EPR, to address these challenges. Specifically, given event streams and scene text information, we first adopt a visual encoder and a text encoder to extract respective feature representations. We then apply optimal-transport constraints to achieve cross-modal alignment. We then design a cross-modal bottleneck fusion module to further fuse bimodal features. Furthermore, we develop multi-scale spatial optimal transport and inter-frame temporal optimal transport modules to mine high-quality spatial-temporal event features for scene recognition. Extensive experiments on the NYC-Event-VPR and EPRBench datasets fully validate the effectiveness of our proposed framework for event-based VPR. The source code of this paper will be released upon acceptance.

0.25

Zero-shot evaluation of open vision-language models as System 1 decision models on radiology tasks

cs.CV cs.AI ·2026-11-01 ·Bradley J Erickson, Sara Salehi
Abstract · 22 pages, 3 tables, 2 figures

Purpose: We compared two multimodal System 1 decision models, Clef 27B and d1 3B, with a traditional LLM (Qwen 3.8-27B) on seven radiology tasks covering text, radiographs, CT and MRI. Methods: Three models were compared: Clef, Qwen3.8-27B and d1. Three tasks were interpretation of text and 4 were detection of finding in images. The tasks were: 1. finding presence of finding in CT-RATE chest CT reports; 2. Presence of finding in PadChest chest radiograph reports; 3. the same 18 findings on CT-RATE chest CT but 3D volumes; 4. MR safety status of devices from AccessGUDID device records; 5. line and tube position on RANZCR CLiP chest radiographs; 6. motion-artifact severity on MR-ART T1-weighted brain MRI and 7. sequence identification on BraTS 2021 brain MRI. The primary endpoint was AUROC. There was no fine-tuning of the models on the tasks. Results: All models completed all tests without a failed request. All retrieved findings from reports almost perfectly (AUROC >0.95). Clef ranked higher than its Qwen backbone on CT-RATE volumes and RANZCR lower on MR-ART. The two did not differ on BraTS, GUDID or PadChest, and differed by 0.002 on CT-RATE reports. On BraTS, Clef correctly identified 90 of 100 FLAIR volumes while Qwen never chose it. d1 ranked lower than Clef on six of seven tasks. No model performed well on motion severity. Clef was faster per request than Qwen on tasks with many questions per input. d1 used 3 to 9 times less energy per request than Clef. Conclusions: System 1 models can perform important tasks in radiology with high accuracy, at lower compute cost than LLMs used in the usual way. When an LLM is constrained to a single answer token, its cost and performance is similar to System 1 models unless multiple questions are asked. A 3B System 1 model was close to the 27B models on report text but not on images. No model performed well on motion severity.

0.25

Vocabulary Aligned Explanations for Computer Vision Models

cs.CV cs.AI ·2026-11-01 ·Erfan Mirhaji, Nikhil Krishnaswamy, Jill Zarestky, Lisa Mason, Nathaniel Blanchard +1 more
Abstract

Optimizing vision models for accuracy does not force the model to learn and utilize human-understandable and domain-specific concepts and rules. We propose explaining image classifications through domain-specific concepts, visual evidence, and expert knowledge expressed as logical rules. We introduce FLUENT (Framework for Localized, Uncertainty-Aware Explanations in Normative Terminology), a post-hoc framework that ranks expert-authored rules combining jointly required, alternative, and absent concepts. To account for prediction errors, FLUENT combines predicted labels and concept states with empirically estimated likelihoods and concept priors. It scores each rule by calculating support over satisfying assignments to the concepts mentioned in that rule, then normalizes these scores to express relative confidence among candidate explanations. The selected rule is illustrated through occlusion-based highlights for positive concept mentions in the query image and localized examples in retrieved reference images for negative mentions. These illustrations are intended to show end users unfamiliar with the domain what the concepts named in an explanation look like. FLUENT combines domain knowledge and model accuracy, and delivers visual concept highlights and explicit rule confidence to explain an image in the vocabulary of an application domain.

0.25

AR-RRG: Anatomy-Guided Retrieval-Driven Radiology Report Generation

cs.CV ·2026-11-01 ·Xiang Zhang, Shijie Song, Wanqing Zhao, Wei Zhou, Rui Cao +6 more
code/data
Abstract

Radiology report generation (RRG) streamlines clinical workflows and alleviates the burden of radiologists' documentation. Standard vision-language models capture global visual cues and lack fine-grained alignment between anatomical regions and diagnostic text. Most retrieval-augmented methods retrieve complete historical reports, introducing redundant noise and severe vision-text mismatches that trigger clinically risky hallucinations and missed lesions. In this paper, we introduce AR-RRG, an anatomy-guided retrieval-driven method for the generation of radiology reports. AR-RRG consists of two core modules: the anatomy-aware self-training lesion detection network (ASLDNet) and the retrieval-driven radiology report generation (RRRG). ASLDNet leverages anatomy-aware spatial verification (ASV) and class-adaptive dynamic thresholding (CDT) via self-training, delivering accurate lesion localization to support subsequent generation. As the generative branch of AR-RRG, RRRG eliminates clinical hallucinations through dual-granularity retrieval (DR) and complements missing lesions with a neighborhood consensus (NC) mechanism. A parallel continuous visual perception (PCVP) stream converts radiographs into visual tokens through fixed medical encoding and resampling. Multi-scale text and visual features are fused, followed by LoRA-efficient large language model tuning to generate clinically consistent, visually grounded radiology reports. Extensive evaluations on MS-CXR, MIMIC-CXR, and IU X-Ray demonstrate that AR-RRG consistently outperforms state-of-the-art methods across mainstream metrics. Our code will be made publicly available.

0.25

Multimodal BVP and Respiration for Automated Pain Recognition with Hankel-FCN

cs.CV ·2026-11-01 ·Valentina Becchetti, Muhammad Umar Khan, Stefanos Gkikas, Alessandro Giuseppi, Raul Fernandez Rojas
Abstract

Automated pain recognition is a fundamental challenge for the advancement of clinical care and empathetic human-computer interaction, due to the subjective and dynamic nature of pain. In this study, we propose a deep learning framework for objective pain assessment based on multimodal physiological signals from the AI4Pain dataset. The approach focuses on Blood Volume Pulse (BVP) and Respiration (RESP), capturing complementary aspects of autonomic responses to nociceptive stimulation. To explicitly model their temporal dynamics, the signals are transformed into structured trajectory representations using Hankel matrix embeddings, enabling the learning of dynamical patterns beyond conventional time-series representations. These embeddings are processed by modality-specific Fully Convolutional Networks, and their outputs are integrated through a late fusion strategy for three-class classification (No Pain, Low Pain, High Pain). Experimental results demonstrate that the proposed method achieves an accuracy of 67.09% and an F1-score of 66.41%, providing competitive performance compared to existing approaches, while relying on no handcrafted feature engineering. Multimodal fusion significantly outperformed either modality alone, and the Hankel embedding showed a trend towards further improvement. Further validation is required to establish generalisability to broader populations and clinical settings.

0.24

SpecUltra: A Unified Foundation Model for Heterogeneous Spectral Images

cs.CV ·2026-11-01 ·Weili Kong, Baisen Liu, Xiaojun Bi, Jiaming Pei
Abstract

Spectral imagery, including hyperspectral image (HSI) and multispectral image (MSI), exhibits substantial heterogeneity in band numbers, wavelengths, spectral resolutions, hindering unified modeling and cross-sensor transfer. From the perspective of imaging physics, these heterogeneous observations can be regarded as discrete measurements of an underlying continuous spectrum acquired through different sampling processes. Motivated by this perspective, we propose SpecUltra, a unified foundation model for heterogeneous spectral images. Specifically, SpecUltra employs 3D convolutional tokenization and wavelength-aware 3D RoPE to encode relative wavelength and spatial relationships, enabling fine-grained spatial-spectral modeling within a shared coordinate system. We further construct MultiHyper-200K, a large-scale multi-sensor HSI dataset, and pretrain SpecUltra using auxiliary-token-assisted mask image modeling (MIM), where a level token indicates the radiometric product type and register tokens absorb sensor- and acquisition-dependent variations. To bridge the spectral-granularity gap between HSI and MSI, we introduce Spectral Kernel Transformation (SKT), which transfers pretrained knowledge in parameter space by constructing MSI embedding kernels from fine-grained HSI-pretrained kernel slices. This design preserves pretrained spectral priors while maintaining compatibility with the pretrained backbone. Experiments on HSI and MSI classification and segmentation benchmarks demonstrate state-of-the-art or competitive performance across disparate sensors and downstream tasks.

0.24

IMPACT: Image-Conditioned Motion with Physics-Aware Collision and Trajectory Guidance for Video Generation

cs.CV ·2026-11-01 ·Shiwei Yang, Yunlong Yu, Zhao Wang, Jungong Han
Abstract

Visually realistic video generation does not guarantee physically consistent motion. Generated objects may exhibit implausible collision responses or rebounds without energy dissipation, while even physically valid motion predictions may not be faithfully realized during synthesis. We propose \textbf{IMPACT}, a physics-state-driven image-to-video framework that addresses these challenges by coupling explicit dynamics prediction with complementary sparse--dense physical guidance. A learnably calibrated Physics State Generator unifies continuous free-flight dynamics with discrete collision responses, linking impact timing, post-impact velocity, rebound attenuation, and eventual settling within a coherent physical-state sequence. To translate this evolution into effective generative constraints, IMPACT derives sparse trajectories and dense region-level motion priors from the same physical rollout. Sparse trajectories constrain global object displacement, while dense priors guide diffusion sampling to encourage coherent object-region motion and suppress unwanted background changes. Their shared physical basis coordinates global and local motion guidance to reduce prediction--generation mismatch. Experiments on 100 controlled sphere--plane collision scenes spanning different material combinations and visual environments demonstrate improved physical plausibility over representative general-purpose, physics-grounded, and motion-controllable baselines. The project repository is available at https://weiweiwiewei.github.io/impact

0.24

Unified Loss-Aware Density Control for 3D Gaussian Splatting

cs.CV ·2026-11-01 ·Tae-Young Kim, Jiwoo Chung, Jae-Pil Heo
top venue
Abstract · Accepted to NeurIPS 2026

The efficiency and fidelity of 3D Gaussian Splatting critically depend on adaptive density control, which grows and suppresses Gaussian primitives throughout optimization. However, the standard pipeline still relies on heuristic view-space gradients, scale thresholds, and global opacity resets, rather than explicitly assessing the loss-reducing effect of each split, clone, or opacity update. This mismatch can allocate primitives to already-sufficient regions, miss under-reconstructed structures, and destabilize training through blanket opacity changes. In this work, we recast adaptive density control as a collection of local loss-reduction decisions. We derive efficient Taylor-based per-Gaussian surrogates that predict the objective change caused by three primitive-level operations: splitting, cloning, and opacity decay. This yields a unified loss-aware framework with three components. First, we extend second-order splitting to a normalized joint position-scale space, enabling children to adapt not only their locations but also their anisotropic extents. Second, we formulate duplicate cloning as a surrogate minimization problem, selecting only Gaussians whose duplicated contribution is predicted to improve reconstruction and assigning each child a principled opacity. Third, we replace global opacity reset with selective soft opacity decay that weakens harmful primitives while preserving useful ones. Across standard 3DGS benchmarks, our method produces more targeted primitive allocation and achieves improved rendering quality with comparable or smaller Gaussian budgets compared to baselines.

0.24

CUFormer: Modality-Aware CT-Ultrasound Point Cloud Registration for Orthopedic Navigation

cs.CV ·2026-11-01 ·Lixing Tan, Hongye Hao, Tianrun Gu, Qing Xia, Yuting Guo +2 more
Abstract

Registering intraoperative Ultrasound (US) bone-surface point clouds to preoperative Computed Tomography (CT) models can support radiation-free orthopedic navigation. Differences in imaging physics produce distinct surface sampling patterns, complicating cross-modal feature matching. Bone strongly reflects ultrasound and casts an acoustic shadow, so US captures only the probe-facing bone surface and leaves most CT points unmatched. This partial coverage makes correspondences ambiguous on near-symmetric bones. We propose CUFormer, a modality-aware registration framework. We introduce a Modality-Aware Transformer (MAT) that encodes input modality through identity embeddings and updates features through modality-specific adapters within a shared geometric matcher. We model imaging visibility and use predicted reference-side visibility to weight coarse and fine correspondences. CUFormer achieves the lowest mean registration errors among evaluated methods on two CT-US datasets, UltraBones100k and UltraBonesHip, and the CT-RGB-D dataset SpineDepth. On UltraBones100k, it reaches mean rotation error 4.53$^\circ$, translation error 3.19\,mm, and Root Mean Squared Error (RMSE) 2.54\,mm, reducing RMSE by 65.8\% relative to the best-performing baseline by this metric. These results demonstrate CUFormer's effectiveness in aligning partial bone surfaces with preoperative CT across anatomical sites.

0.24

TrackMem: Tracking Evolving Beliefs in Long-Horizon Financial Dialogue

cs.CL cs.AI ·2026-11-01 ·Austin M. Y. Cheung, Harsh Vardhan Gupta, Go Nam Lui, Lok Kan Chan, Ginny Y. Wong +1 more
benchmark
Abstract · 22 pages (6 pages main text), 4 figures

Memory systems for LLM agents are evaluated mostly on social dialogue, where facts accumulate and rarely change, so remembering reduces largely to finding the turn in which something was said. Financial dialogue breaks this assumption: figures are revised, stances shift without being restated, rules stated once govern later questions, and a single company is discussed under several analytical lenses. Memory systems that retrieve the passages most similar to a question, or the most recent ones, therefore return superseded values, miss rules that share no wording with the question, and cannot recover a stance spread across sessions. We present TrackMem, a per-speaker graph memory that tracks the history of figures, separates perspectives by entity and lens, and preserves each speaker's actions in chronological order. To evaluate it, we introduce FinLoCoMo, a benchmark of 15 long-horizon investment conversations with 765 questions in two splits: one following the LoCoMo categories and one targeting four failure modes. With two readers, TrackMem outperforms existing memory systems on both splits. On the failure-mode split it reaches 73.9% macro accuracy against 69.4% for the strongest baseline, with the largest gains on questions about revised figures (86.5% against 73.4%).

0.23

Vision-language classification of Torsades de Pointes risk from pseudo-coloured ECG images: do T-wave morphology and colour context help or hinder?

cs.CV ·2026-11-01 ·Dario Labrador Alonso, Alaa Alahmadi
Abstract

Contextual prompting is widely presented as a cheap route to better performance from vision models, but the evidence comes largely from benchmarks in which an attribute named in a prompt is also present in the image, confounding the two channels. We investigate the significance of context for a semantically complex biomedical signal classification task using a domain-adapted vision-language model, BiomedCLIP. Building on our previous work that renders electrocardiogram (ECG) signals as colour-enhanced images to flag Torsades de Pointes risk using the QT-nomogram and pseudo-colouring technique, we cross four context strategies with four T-wave morphology label sets, evaluated zero-shot and with a class-balanced linear probe, under five-fold stratified cross-validation on 9,789 images. Zero-shot inference is unusable. Once trained, the colour-enhanced images outperform the colourless ones on every metric: balanced accuracy 0.903 versus 0.825, sensitivity 0.914 versus 0.827, specificity 0.892 versus 0.823, ROC-AUC 0.955 versus 0.888 and Matthews correlation coefficient (MCC) 0.437 versus 0.311. Crossing colour-in-image with colour-in-prompt separates the confounded channels: the image contributes +0.126 MCC (+0.078 balanced accuracy, +0.067 ROC-AUC) whether or not the prompt names colour, whereas naming it changes nothing under any metric. Context is two-sided: the best prompt set gains 0.008 MCC where colour is present and 0.016 where it is not, while the worst loses 0.017 and 0.013. For a signal rendered as an image, the leverage lies in how the signal is drawn rather than in how it is described - and where the endpoint is an interval measurement, prompts naming T-wave morphology describe the wrong quantity.

0.22

Learning-to-Memorize: Dynamic Context Management for Long-Horizon Autoregressive Video Generation

cs.CV ·2026-11-01 ·Haowei Zhu, Qijie Wang, Jia Li, Xing Wang, Tianyu Zhao +5 more
top venue
Abstract · Accepted at NeurIPS 2026

Autoregressive video generation has achieved impressive results on short clips, yet generating long videos remains challenging due to error accumulation over extended horizons, where each predicted frame depends on potentially imperfect previous frames. Existing approaches typically rely on static context management strategies, such as sliding-window caches that discard early frames or fixed sink frames that permanently preserve initial content. These designs either lose critical long-range context or anchor generation to outdated information, which can lead to degraded fidelity, motion stagnation, and reduced diversity. In this paper, we propose Learning to Memorize (L2M), a learned context memory management framework that dynamically retains, evicts, and evolves historical frames according to predicted importance. The framework consists of an importance prediction router that estimates the relevance of each historical frame, degradation-aware training that improves robustness to noisy or corrupted context, and a dynamic memory initialization and evolution mechanism that updates long-term memory using exponential moving average (EMA) importance scores. This allows high-quality frames to replace stale anchors and enables memory to adapt naturally to scene evolution. Trained with a teacher-student distillation objective and sparsity regularization, L2M makes efficient use of a fixed KV-cache budget while preserving the most informative contextual information. Extensive experiments demonstrate that L2M improves long-horizon video generation quality and temporal consistency over existing baselines, providing a learned strategy for context memory management in long autoregressive video generation.

0.22

CULTURE-CMT: Benchmarking Contextualized Social Media Translation

cs.CL ·2026-11-01 ·Linjuan Wu, Dingming Li, Yiwen Qiu, Xinze Lyu, Ye Guo +3 more
benchmark
Abstract · 24 pages. Preprint

Social-media comments are interactions, yet they are often translated and evaluated as isolated text. This mismatch is especially problematic when meaning depends on surrounding discussion, community-specific slang, or culturally situated expressions. We introduce CULTURE-CMT, a Chinese-English benchmark for contextualized social-media comment translation that restores the post and conversational context naturally available to users. Its 572 manually verified examples are organized into Slang, Context, and Both, enabling diagnostic analysis of distinct and interacting sources of translation difficulty. To support scalable evaluation, we further develop a retrieval-augmented cultural-effectiveness Judger that combines a calibrated scoring model with dynamically retrieved cultural knowledge. CULTURE-CMT reveals several non-trivial patterns. Providing interaction context improves most model configurations, but does not eliminate culturally ineffective translations. Examples requiring both contextual grounding and culturally situated language remain substantially more difficult than either source of difficulty alone. Moreover, larger models are not consistently better, and thinking produces highly model-dependent gains, including occasional degradation. In contrast, a 27B model adapted on benchmark-disjoint social-media data achieves the lowest observed invalidity rate among the evaluated configurations under thinking inference with context. Together, these results suggest that realistic social-media translation is not simply a matter of scaling models or reasoning longer, but of learning to translate meaning as it is constructed within evolving online interactions.

0.22

AbacusGS: Gaussian Splatting without Per-Splat Sorting

cs.CV cs.GR ·2026-11-01 ·Jiacheng Ding, Yanliang Qi, Cong Guo, Peihua Mai
Abstract · 10 pages, 5 figures, 6 tables

3D Gaussian Splatting sorts every splat by depth in every frame. Viewers that draw splats with the hardware rasterizer and sort them on the CPU, as most web and mobile viewers do, spend most of the frame time in this sort. We present AbacusGS, which stores the blending order of a 3DGS scene in the scene layout. AbacusGS groups the splats into octree boxes of at most 128 splats and stores each box in order along an axis chosen at write time. In each frame, AbacusGS sorts only the visible boxes, and one dot product per box decides whether the box is read forward or backward. A short order-aware fine-tuning converts an existing 3DGS model, and the over operator of 3DGS stays unchanged. On 13 standard scenes, AbacusGS loses 0.13 dB PSNR on average against an equally fine-tuned 3DGS model. In an OpenGL viewer that orders splats on the CPU, AbacusGS renders frames 4.5 times faster than per-splat sorting. We also present a compact format of 17 bytes per splat and a CUDA tile renderer that use the same order. The code will be released.

0.22

Acoustic and Attentional Challenges Shape the Balance Between Attended and Competing Speech Tracking Differently

cs.SD eess.SP ·2026-11-01 ·Johanna Wilroth, Martin A. Skoglund, Emina Alickovic, Martin Enqvist
Abstract

Selective attention enhances cortical tracking of attended speech, but the role of ignored speech remains less clear. In particular, it is unknown whether attended and ignored speech contribute to cortical tracking in a fixed balance or whether this balance changes with listening conditions. We addressed this question across six electroencephalography (EEG) datasets spanning acoustic and attentional challenges, hearing impairment, and hearing-aid processing. We systematically varied the relative weighting of attended and competing speech and studied how their combined contribution affected cortical speech tracking. Across most listening conditions, cortical tracking was improved when we included competing speech, compared with attended speech alone. Importantly, the balance between attended and competing speech changed across conditions. Increasing acoustic difficulty increased both the benefit and the optimal contribution of competing speech, with the same general pattern observed in normal-hearing and hearing-impaired listeners. In the hearing-aid datasets examining noise-reduction conditions, the contribution of competing speech was more limited. Attentional challenges also changed the contribution of competing speech, but differently from acoustic challenges. These findings show that attended and competing speech do not contribute to cortical tracking in a fixed balance. Rather, their balance depends on the listening challenge. Although these results do not define a single optimal ratio, they provide a step toward hearing devices that adapt not only to the attended talker, but also to how competing speech is balanced in complex listening environments.

0.22

EvoRec: A Unified Self-Evolving Agentic Recommender with Dynamic Reflection Trees

cs.IR ·2026-11-01 ·Mengting Li, Long Zhang, Hao Jiang, Sheng Yu, Fei Pan +1 more
Abstract

Conventional recommender systems rely on fixed cascaded pipelines, limiting their ability to adapt to evolving user needs and feedback. Existing agentic recommenders introduce reasoning and tool use but often lack end-to-end decision control and systematic failure reflection. We propose EvoRec, a unified agentic recommendation framework that combines autonomous decision-making with reflection-driven self-evolution. A Need Agent infers user needs from long-term, short-term, and working memory. A Planning Agent adaptively allocates retrieval budgets between Behavior Continuation and Need Satisfaction, selects retrieval tools, and sets ranking-fusion weights. An Acting Executor executes the plan and records decision traces. Following user feedback, a Dynamic Reflection Tree identifies failures in need inference, planning, or previous repairs, proposes personalized corrections, and validates them through replay. Verified repairs update working memory and are distilled into reusable cross-user strategies, enabling continual improvement without parameter updates. Experiments on a Kuaishou local-services advertising dataset show a 15.7% Hit@50 improvement over the strongest baseline. Hierarchical reflection further improves Hit@50 by 13.1% among users with prior repairs. In a seven-day online A/B test, nearline EvoRec increases advertising revenue by 4.6% and CTCVR by 3.5%, with a 14.5% revenue gain for high-intent users. These results demonstrate the effectiveness of reflection-driven self-evolution in industrial-scale recommendation.

0.22

VLMs Read Aloud: The Gap Between Recognizing and Using Visual Text

cs.CV cs.CL ·2026-11-01 ·Zhiyun Shi
Abstract · 23 pages, 7 figures, 21 tables

Visual-text compression renders long contexts as images and is engineered around how much of the text a vision-language model (VLM) can recover. Yet recovered text is not usable text. VLMs can transcribe the text in an image word for word and still fail to answer from it; given back their own transcript, they answer nearly as well as from the original text. We call this difference the read-use gap. Recognizing and using visual text have different resolution requirements: Qwen3-VL-8B's recognition saturates at 15-pixel characters while use catches up only at 24 pixels, and scaling from 2B to 32B raises recognition without narrowing the gap. Putting the evidence into words closes it: thinking VLMs recite the evidence in their thinking, and their own transcript, written without seeing the question, recovers most of what thinking adds. In a preregistered held-out test, the gap ranks how much eight VLMs gain from their own words over pixels (Spearman $ρ$ = 0.86), which recognition rates do not. We propose ALOUD, which spends visual resolution on finding and recognizing evidence and language tokens on using it. Without keeping the source text, it doubles the accuracy of a frozen Qwen3-VL-8B on long documents compressed 5x and 10x: at 5x, from 27.7% to 54.8%, near the original text's 58.5%, with a context smaller than the compressed pages themselves. For less than the cost of reading the original text, it leads at equal budget every method that does without it, at every level, and beats AGAR and VERA, which paste the original text back; at 5x it also beats LensVLM-9B, which is trained and keeps the original text. Visual-text compression is limited not by how much text a model can recover, but by how much it can use.

0.22

Write Facts the Way Models Read Them: Lifting the Two-Hop Curse

cs.CL cs.LG ·2026-11-01 ·Zhiyun Shi
benchmark
Abstract · 17 pages, 4 figures, 16 tables

A language model that learns a new fact by fine-tuning can repeat it but cannot reason with it: taught that a person is a citizen of a new country, it still cannot say, without chain-of-thought, who the head of state of that person's country is. This two-hop curse is widely treated as a reasoning deficit, and the strongest knowledge editors answer it with synthesized reasoning data. We show that the reasoning is intact and the writing is wrong. Fine-tuning writes a new fact as an answer to be said, whereas the second hop reads the bridge ("that person's country") as an entity, from the middle layers on: patching the model's own representation of the entity into the bridge at those layers largely restores the second hop, yet after fine-tuning the bridge there is no closer to the new entity than before the edit. We therefore propose read-equivalent writing (Legible): one loss term makes the bridge phrase, at the layer where the second hop finishes reading it, match the model's own reading of the new entity, and keeps the same relation's other subjects unchanged. Phrases and targets come from the edited model itself, and no multi-hop training data is used. Across Qwen2.5, Llama-3 and Qwen3, Legible (I) composes learned facts more often than the model composes its own knowledge, on MQuAKE's two-hop-curse questions; (II) reaches 1.8-2.6x the bridge accuracy of CaKE, whose reasoning data composes no better than plain fine-tuning, and has the best average over composition, accuracy and locality against four families of writers; and (III) lifts the curse on RippleEdits, on the benchmark of Balesni et al., who named the curse, and at 14B.

0.22

Consolidating Reasoning with Test-Time Training

cs.CL cs.AI ·2026-11-01 ·Haotian Wu, Zeming Chen, Hao Zhao, Antoine Bosselut
published
Abstract · NeurIPS 2026, 25 pages, test-time training, reasoning aggregation, meta-learning

Effective multi-agent collaboration requires finalizing solutions from reasoning that is distributed across the individual attempts of many agents. However, useful intermediate reasoning can remain unexploited when collaboration is limited to voting on conclusions reached among individual agents, or aggregating long, noisy collections of reasoning trajectories in context. In this work, we propose TTT-Consolidate, a framework that uses multi-agent reasoning trajectories as adaptation data to improve a model's subsequent reasoning. TTT-Consolidate jointly consolidates trajectories into parametric representations through test-time training, then generates a new solution by conditioning on the consolidated information. We meta-learn how to consolidate using bi-level optimization at training time, where an inner loop encodes reasoning trajectories into a lightweight LoRA adapter, and an outer loop learns meta-parameters for the inner loop such that the consolidation is helpful for the model to synthesize a final solution. As no large-scale corpus of diverse reasoning trajectories exists for training aggregation methods, we construct the OpenParallelThinking corpus composed of 12,810 math problems paired with a mixture of trajectories from six LLMs. Evaluated on challenging math and STEM benchmarks, TTT-Consolidate outperforms the best aggregation baselines by up to 20 percentage points, even under matched compute and runtime settings, with the gap widening at larger model scales. TTT-Consolidate also recovers from unsuccessful individual trajectories more robustly, an ability that improves as the number of available (even incorrect) trajectories increases. Finally, we demonstrate that TTT-Consolidate benefits from both SFT and RL-based outer loop objectives. We will release TTT-Consolidate and OpenParallelThinking for use in future work on reasoning aggregation.

0.21

Conditional Diffusion-Based Generation of T2-Weighted Prostate MRI for Clinically Significant Prostate Cancer Classification

cs.CV ·2026-11-01 ·Xavier Rafael-Palou, Alejandro Vergara-Richart, Ignacio Iborra Roncales, Carmen Prieto-de-la-Lastra, Andreu Antolin-Redondo +2 more
Abstract

Scarce and unrepresentative training data limit the performance of prostate MRI prediction models. Synthetic T2-weighted (T2W) MRI may augment training data containing clinically significant prostate cancer (csPCa) and transition zone (TZ) lesions. In this retrospective, single-center, ethics-committee-approved study, we developed a conditional generative model for T2W prostate MRI extending MAISI-v2 with csPCa class, zonal, and lesion-mask-guided conditioning, incorporating a relative advantage weighted loss fine-tuned on 1,303 patients (1,654 lesions). Fidelity was assessed by incremental ablation, and downstream utility was evaluated with T2W-based csPCa classifiers across five simulated training-data scenarios, each testing 18 synthetic-data configurations. The proposed model reduced Fréchet inception distance (FID) from 13.28 (pretrained MAISI-v2) to $2.04 \pm 0.06$ internally (PI-CAI: $3.83 \pm 0.51$). Synthetic augmentation significantly increased AUC in all five scenarios (+0.027 to +0.088; adjusted $p = 0.016$--$0.043$), though cross-institutional augmentation of PI-CAI yielded a non-significant gain (+0.013; $p = 0.397$). Overall, our model improved T2W MRI fidelity, and synthetic images improved csPCa classification under simulated data constraints, although benefit did not transfer significantly across institutions.

0.21

Scaling Makes It Possible: How Large Models Master Impossible Languages

cs.CL cs.AI ·2026-11-01 ·Yaning Jia, Chunhui Zhang, Chiyu Ma, Zhongyu Ouyang, Ivory Yang +1 more
Abstract · 17 pages, 7 figures, including 6 pages of supplementary information

While language models (LMs) excel in natural language tasks, their ability to learn impossible languages--those that violate the structural, cognitive, or logical constraints that make human languages learnable and functional--remains debatable. This controversy is further fueled by a divide between linguistic theorists and empirical researchers working with small-scale LMs. To address this controversy, we introduce semantic coherence. This concept measures how effectively semantic information is expressed in language structures and examines the learning capabilities of LMs across multiple scales. By fine-tuning billion-parameter LMs on impossible language samples, we demonstrate that larger LMs achieve similar performance on impossible and possible languages, whereas million-parameter LMs exhibit significant degradation when trained on impossible languages. This performance gap widens as semantic complexity increases. We further show that scaling up the LM size fundamentally shifts the learning focus from grammatical rules to semantic understanding, with billion-scale LMs exhibiting a threefold reduction in sensitivity to low-level linguistic perturbations while maintaining stable performance under high-level semantic perturbations.

0.21

Prior-Guided Physics-Informed Deep Learning for Fast 4D MRI Reconstruction

cs.CV physics.med-ph ·2026-11-01 ·Mojtaba Safari, Parag J. Parikh, Shansong Wang, Mingzhe Hu, Philip Connell +4 more
Abstract

Background and purpose: We propose PGPI4MR (prior-guided, physics-informed 4D MRI reconstruction), a deep learning method that leverages information from other respiratory phases or a setup scan, the first 30 s of the same scan, to improve 4D MRI reconstruction without extra scan time. Materials and methods: PGPI4MR is an unrolled network that reconstructs 16 respiratory phases from 60, 30, or 15 s of data, and single 3D volumes, each from 0.29-0.49 s of data, using a setup-scan image. The networks were trained once, on other volunteers, and were not retrained for new volunteers. We evaluated PGPI4MR on measured k-space from three datasets: free-breathing MR-linac (1.5T) and abdominal (3T) scans, and thigh scans with cuff compression (1.5T). All methods, including comparative models, received identical samples and were scored similarly. Results: In vivo, PGPI4MR had the lowest held-out error for 4D MRI at every scan time, and its 16 phases were ready a median of 7.5-13 s after the scan on the 2 GPUs tested. Coupling the 16 respiratory phases reduced the error at 60 s by 9.1% compared with reconstructing each phase alone. For sub-second MRI, the error at 0.49 s was 15.5% lower than that of the best comparison method, classical reconstruction toward the same prior, and on the faster GPU, every 0.29 s window was acquired and reconstructed within 0.5 s. Without retraining, PGPI4MR matched the best methods on the 3T abdominal scans but not on the thigh scans; after retraining on thigh data, it reduced the held-out error by 6.6% relative to the best comparison method. Conclusion: Prior information allowed PGPI4MR to reconstruct 4D and sub-second MRI more accurately than compressed sensing and a published 4D network, within seconds and without extra scan time. This could support 4D MRI for treatment planning and 3D motion monitoring during MR-guided radiotherapy.

0.21

Refining Roadside 4D Radar Vehicle Classification Using Point-Cloud Dispersion

cs.CV ·2026-11-01 ·Mohammed Alsakabi, Andreia Figueiredo, António Almeida, John Dolan, Miguel Luís +3 more
Abstract

Roadside sensing infrastructure is essential for Intelligent Transportation Systems (ITS), but cameras and legacy sensors are limited in range, weather robustness, and depth accuracy. 4D radar complements these modalities by resolving range, azimuth, elevation, and radial velocity under adverse conditions, though its sparse, noisy point clouds make reliable classification difficult. We present a roadside multi-sensor platform combining a 4D radar and stereo depth camera under synchronized acquisition, used to collect a real-world dataset. Our analysis reveals a substantial class-distribution discrepancy between modalities, most notably a strong radar confusion between passenger cars and motorcycles (49.01% and\ 48.60% of the dataset). To address this, we propose a Neyman-Pearson-based binary classifier on the temporally-smoothed 2D positional standard deviation of radar point cloud clusters, reducing classification error from 52.1% to 42.7%. This lightweight and edge-deployable approach improves radar-only classification reliability and lays the foundation for future camera-radar fusion.

0.20

Information-Theoretic Design of Sparse Autoencoder-Based Steering

cs.CL cs.LG ·2026-11-01 ·Hirofumi Morise
published
Abstract

Sparse autoencoder (SAE)-based activation steering is a lightweight approach for modifying the behavior of large language models (LLMs) by adding small vectors to hidden states during inference. However, designing effective steering interventions remains challenging. In this work, we develop an information-theoretic framework for analyzing SAE activations through empirical feature-firing distributions. The central object of the framework is the Jensen--Shannon divergence (JSD) between such firing distributions, which quantifies how far a generated or intervened state deviates from a dataset-induced target distribution. Based on an analytic divergence response for virtually shifted states, we construct steering vectors toward the target firing distribution. The same divergence can also be used to monitor deviation from another empirical distribution, such as a general valid-language distribution. We demonstrate the effectiveness of the steering in a question-mode generation task study. We further use teacher-forced pulse experiments to show that local interventions induce structured downstream JSD responses in later layers, and apply the framework to a safety-related target--contrast response setting for feature analysis.

0.20

Modeling Visual Cortex with Layerwise Multiscale Manifold Capacity Maximization

cs.CV q-bio.NC ·2026-11-01 ·Thomas Yerxa, SueYeon Chung, Eero P. Simoncelli
Abstract

Task-optimized deep neural networks have risen to prominence as the most predictive phenomenological models of responses in primate visual cortex, but leave much to be desired from the perspective of biological plausibility. One such limitation is the reliance on precise credit assignment through global backpropagation of error signals. Recent work has shown that this weakness can be circumvented by requiring each subsequent stage to solve a distinct and increasingly complex task, allowing for layerwise local learning signals. We propose a novel strategy for crafting such intermediate losses that uses an efficient coding framework formulated in terms of manifold capacity, and which can be computed using a sequence of {canonical cortical computations}. In particular, we leverage the relationship between the multiscale nature of visual signals and the dilation of receptive field sizes in cascaded visual representations to modulate complexity, allowing for the reapplication of these common loss computations at each stage of the hierarchy. We evaluate our approach on its ability to predict neural datasets spanning three areas of the ventral stream hierarchy in macaques, as well as human psychophysical data on an object classification task. We find that our unsupervised layerwise model matches or exceeds the performance of competitive architecture-matched baselines on all evaluations considered.

0.20

See, Think, Use: Diagnosing Perception and Reasoning Failures in VLMs

cs.CV cs.AIcs.CL ·2026-11-01 ·Jiacheng Ding, Yanliang Qi, Cong Guo, Peihua Mai
Abstract · 10 pages, 4 figures, 8 tables

We present See, Think, Use, a protocol that traces each answer of a vision-language model to perception, reasoning or the use of perception. The diagnosis shows a developer which stage to improve and directs test-time compute to the stage that fails. Benchmarks score the final answer, and an answer that rests on the image earns the same credit as a guess or a memorized prior. The protocol asks the model two extra questions per item and needs no access to its weights. A localization probe asks for a box around every instance and measures seeing, and a text-replacement probe answers from a description and measures thinking. Use is the rate at which the answer follows the model's own perception across counterfactual image pairs. The rate needs no answer labels, because the question and its prior are the same on both images. On two model families and six datasets the protocol shows that the knob that helps follows the task family. Answers follow the model's perception except on images that contradict a common prior, and asking the model to localize first repairs this failure when its boxes are right. Allocating compute by the measured stages comes close to the best uniform policy at lower cost and beats a random allocation of the same compute wherever a knob raises the short stage. We will release the code, the prompts and the per-question outputs.

0.20

Learning to Look: A Foundation for Visual Understanding

cs.CV ·2026-11-01 ·Wenshuai Zhao, Yihao Wang, Benjamin Robson, Juho Kannala, Arno Solin
Abstract · 15 pages, 7 figures

Vision-language models (VLMs) suffer from visual hallucination, such as saying an object that is not present in an image. We ask a more fundamental question: whether a VLM is even aware of where it is looking before understanding the semantic meaning in the image. We formulate this question as a LOOK task: identifying which parts of its visual input were available to the model. We construct the task in a self-supervised way without manually labeling the available region to the model. Our experiments show that existing VLMs fail to localize the visible region. We further show that fine-tuning VLMs on the proposed LOOK task improves a model's awareness of visual input.

0.20

SurgStream: A Workflow-Anchored Proactive Streaming Framework and Benchmark for Real-Time Surgical Video Understanding

cs.CV cs.AI ·2026-11-01 ·Ruipeng Ma, Qing Xu, Jie Liang, Zhen Chen
benchmark
Abstract

Surgical video understanding is essential for intraoperative assistance such as workflow monitoring, instrument preparation and procedural documentation. Delivering such assistance in real time requires a model to jointly perceive the ongoing scene, remember past surgical events, and respond before clinically relevant moments pass. Existing approaches provide limited support for jointly retaining procedural evidence and deciding when to respond in streaming surgical video. In this work, we propose SurgStream, a workflow-anchored proactive streaming framework for real-time surgical video understanding. We first introduce Workflow-Anchored Memory (WAM) to record structured instrument-action-target observations and fold them into phase-level summaries, providing compact and queryable surgical memory. Deadline-aware Transition Learning (DTL) is then devised to supervise response states within clinical windows, enabling timely responses before procedural deadlines. We further introduce SurgStream-Bench, comprising 11.7K streaming interactions across various tasks covering real-time perception, backward memory queries, forward proactive QA, proactive alerts and caption memory. Experiments on SurgStream-Bench show gains over state-of-the-art methods, especially in memory and forward proactive QA. The source code will be available.

0.20

Online Active Learning Architecture for Safety-Critical Control Systems

cs.CV ·2026-11-01 ·Shuiwen Shen, Jinhua Zhang, Yan Wang, Huanqiang Zeng, Yueyun Xu +3 more
Abstract · 16 pages, 10 figures

The transition of information-centric artificial intelligence to safety-critical applications is impeded by the limitations of conventional offline-training paradigms in open-world environments. This paper presents Control AI, an architecture for Online Active Learning (OAL) inspired by the continuous Prediction-Evaluation-Correction process of biological intelligence. The framework combines the hierarchical representation capacity of deep neural networks with the real-time error-feedback adaptation of adaptive control. A prediction model maps sensory input into a latent manifold, an optional latent model represents transitions in that space, and a reconstruction model maps the latent state back to observable signals. Residuals between predictions and measurements provide the internal learning signal. Spatial and temporal cross-reconstruction architectures prevent OAL from collapsing to a trivial identity mapping, while coupled, dual, and localized variants shorten error-transmission paths. Localized predictive coding style OAL is implemented through guidance-backpropagation and target-backpropagation architectures, and hybrid and scaled variants support high-dimensional deployment. The paper focuses on the architectural rationale and the progression from a conceptual feedback loop to practical large-scale configurations.

0.20

HaloGuard 2.0: An Open-Weights Constitutional Output Guard for Multilingual Response Safety

cs.CL cs.CR ·2026-11-01 ·Ashmiya Lenin, Navaneeth Sangameswaran, Preetham S
Abstract · 23 pages, 6 figures, 16 tables

An input guard decides whether a request reaches the model, but never sees what the model says. Harm that emerges during generation, because a jailbreak passed the input layer or a benign request drifted into operational detail, is visible only on the output side and only as tokens arrive. We present HaloGuard 2.0, an open-weight streaming output guard that assigns a safe/unsafe probability to every response token in a single forward pass, so a deployment can stop the stream once harm is confirmed. The guard reuses the constitution of HaloGuard 1.0, our input guard, to organise a 478,532-response training corpus, and supervises each token from the point where the response becomes harmful. We find that a guard trained on conventionally shaped data learns a shortcut specific to streaming: it judges the prompt instead of the response, and handles only 13.9% of matched safe/unsafe response pairs correctly. We correct this with a second phase on 106,043 counterfactual responses built around the base guard's own failure modes, and release the equal-weight average of the base and corrective models. On six English response-safety benchmarks, HaloGuard 2.0-0.8B reaches a macro F1 of 83.6 (83.3 on held-out halves of the decision-rule search), up from 72.8, with 38% fewer false positives. It matches the strongest full-response guards with 5-10x fewer parameters and is 2.6 points above the strongest streaming guard, Qwen3Guard-Stream-4B. A multilingual variant, HaloGuard 2.0-Poly-0.8B, trained on PolyGuardMix, reaches 83.1 F1 on the response labels of PolyGuardPrompts across 17 languages, 5.0 points above the strongest baseline. On one H100 the guard costs 0.37 ms per token per stream when moderating 64 concurrent responses. Both guards are hardened continuously through the weekly red-teaming protocol of HaloGuard 1.0, extended here to the output layer.

0.19

High Frequency Safeguard and Weighted Objective Scoring Bandwidth Extension

cs.SD eess.AS ·2026-11-01 ·Tarikul Islam Tamiti, Sajid Fardin Dipto, Anomadarshi Barua
Abstract

Speech Bandwidth Extension (BWE) reconstructs missing high frequencies from a band-limited signal. Please note that perceptually important information is often concentrated in short high-frequency (HF) events. Therefore, an effective BWE requires both accurate HF generation and explicit control over where, how often, and in what spectral form HF energy should be produced. To address the above HF supervision, we present HFSWO, a compact complex-spectral Swin-based dual-stream adversarial framework with two newly proposed specialized discriminators: 1. High-Frequency Safeguard Discriminator (HFSD) converts excessive framewise HF-energy ratios into a differentiable safeguard-constraint penalty optimized through a projected primal--dual augmented-Lagrangian update; 2. Weighted Objective Scoring Discriminator (WOSD) learns an interpretable convex objective scoring over spectral flatness, centroid, and rolloff. Our rigorous experiments on VCTK English and MLS French datasets across multiple frequency ranges, including noisy conditions, show significantly improved perceptual scores with 74.6% fewer parameters than the existing baseline AP-BWE, establishing a new baseline in the BWE task in multilingual and multispeaker setups.

0.19

RIFT: Residual-Informed Flow Transfer for One-Step Generative Video Compression

cs.CV ·2026-11-01 ·Qizhe Wang, Jiaye Fu, Feng Ye, Haocheng Tang, Zetian Song +3 more
code/data
Abstract

Recent advances in video generation have enabled generative video compression to reconstruct details lost during coding using pretrained spatiotemporal priors. However, a substantial gap remains between codec-induced degradation and the perturbation trajectories learned by flow-matching models. In this paper, we identify a compression-dependent directional mismatch that degradation-strength-based timestep calibration alone cannot resolve. Compressed latents remain strongly aligned across codec operating points, while their reconstruction residuals vary substantially in magnitude and direction. To bridge this gap, we propose RIFT, a Residual-Informed Flow Transfer framework for one-step generative video compression. At its core, RIFT introduces a Residual Geometry Inducer (RGI) that converts the pretrained flow prediction into a content-adaptive restoration update using decoder-visible codec cues. RGI integrates flow-timestep conditioning with a learned full-band residual-strength estimate and content-conditioned direction and gain modulation, aligning the orientation and magnitude of the update with the compression-dependent residual geometry. Extensive experiments on three public video compression benchmarks show that RIFT outperforms competing generative video compression methods. The code is publicly available on RIFT.

0.19

Teleoperation System for Service Robots using Virtual Reality Headset and Pose Estimation

cs.CV ·2026-11-01 ·Eduardo Fernandes, Tiago Ribeiro, Antonio Ribeiro, Carolina Lopes, Gil Lopes +1 more
Abstract · 17 pages, 16 figures

This paper presents a teleoperation system for service robots that integrates human pose estimation with virtual reality (VR) to enable intuitive remote control. By using the Mediapipe framework for 2D pose detection and extending it with RGB-D data from the Intel RealSense D455 camera, a comprehensive 3D human pose estimation is achieved. This data is then mapped to the robot's kinematics, allowing the operator to control the robot's movements naturally. A VR interface was developed to provide an immersive perspective, enhancing the operator's situational awareness. Real-world experiments were conducted using the CHARMIE service robot to validate the system's effectiveness in tasks involving object manipulation. While the current system successfully demonstrates the potential of VR-based teleoperation, challenges remain in accurately capturing certain joint angles (e.g., wrist yaw) and ensuring precise mapping between human and robot movements. Future work includes the integration of inertial measurement units (IMUs) for improved rotational tracking and the implementation of more advanced control algorithms. Overall, this research contributes to the development of more intuitive and effective teleoperation systems for service robots, facilitating their deployment in complex and dynamic environments. This work further introduces a novel approach to teleoperation by combining RGB-D sensing with an innovative algorithm for real-time joint angle computation using Mediapipe, enhancing the precision of human-robot motion mapping and enabling immersive control through a VR interface. These contributions represent significant advancements in the field of human-robot interaction, particularly for service robot applications.

0.19

Same Trajectory, Different Knowledge: Information-State-Aware On-Policy Distillation for Search Agents

cs.IR ·2026-11-01 ·Minhao Sun, Xian-Ling Mao, Eric Tao, Zi-Ao Ma, Zhijing Wu
Abstract

On-policy distillation trains search agents with teacher feedback on student-generated trajectories. Recent methods augment the teacher's context with supporting evidence, reference answers, or hindsight information. However, this privileged information may allow the teacher to rely on facts that the student cannot reliably recover from its current state. The resulting information-state mismatch can lead the teacher to undervalue retrieval that helps the student establish those facts. To address this mismatch, we introduce Information-State-Aware Distillation (ISA-Distill), which adapts teacher feedback to the student's factual recoverability. ISA-Distill estimates how reliably the student can recover action-relevant knowledge and derives teacher feedback for the same sampled action from two complementary views. The fact-use view supplies the teacher with the relevant fact, producing feedback that guides the student in using knowledge. The evidence-seeking view instead presents the information need, producing feedback that guides the student in acquiring supporting evidence. The two signals are weighted toward fact use when recoverability is high and toward evidence seeking when it is low, and the resulting feedback guides policy optimization. Experiments across diverse QA benchmarks and model scales demonstrate improved answer accuracy and retrieval efficiency. Our code and model checkpoints will be open-sourced.

0.18

GatedVLA with Camera-Based Emergency Braking: AlpaSim E2E Closed Loop Challenge Technical Report

cs.CV cs.RO ·2026-11-01 ·Kaneyoshi Hiratsuka, Naoki Takada, Makoto Kawano
Abstract

This report presents GatedVLA, our entry for the Physical AI AV (PAI-AV) track. Our learned vision-language-action (VLA) policy pairs the driving-pretrained Qwen-Drive-1.0-4B vision-language model with a layer-wise flow-matching action expert that outputs a 2.5 s horizon of acceleration and curvature controls, augmented with a soft visual mask applied to the expert's cross-attention. Compared to an identically trained baseline, the masked policy achieves comparable overall open-loop error while reducing error on left turns. In local closed-loop evaluation across the same 150 scenes (two rollouts per scene), the visual mask increases the mean scene score from 0.479 to 0.517 (paired difference +0.038, 95% bootstrap CI [-0.018, +0.093], wins/losses 37/26, sign test p = 0.21). Incorporating a rule-based automatic emergency braking (AEB) module at deployment further increases the mean score to 0.599 (+0.082, CI [+0.023, +0.143]) and reduces the mean number of scenes with at-fault collisions from 34.0 to 11.0.

0.18

XDecideBench: Evaluating Typed Decision Models and Confidence Policies Across Languages

cs.CL ·2026-11-01 ·Marek Šuppa, Jaroslav Kopčan
benchmark
Abstract

Typed Decision Models such as Jev, Clef and pplx-decider return a probability for each answer an application provides. Applications use them in many languages, but are compared mostly in English. We introduce XDecideBench, a multilingual benchmark of nine suites in 122 languages. On our twelve-language suite, three open models fine-tuned for typed decisions beat their Qwen base models in every language. On new test data with local base models, their accuracy gain is larger in languages with less web text than in those with more. They fix a similar share of base-model errors in both, so the larger gain may reflect weaker base models. The best model on the twelve-language suite is best on only two other suites. The best English model on each suite, chosen in hindsight, is significantly beaten in about a quarter of suite and language combinations. On Belebele, Decision-1 is best in English but falls far behind Jev in Burmese. A confidence threshold fitted on English to allow 5% errors gives a higher error rate on XNLI, even in German and French. XDecideBench shows that results from one suite or from English are a poor guide to how a model and its threshold behave in other languages.

0.18

EAGR: Evidence-Guided Adaptive Graph Retrieval

cs.IR ·2026-11-01 ·Sonam Gupta, Vignesh P, Vishwajeet Kumar, Adnan Qidwai, Anand Eswaran +2 more
Abstract

Retrieval-Augmented Generation (RAG) for multi-hop reasoning requires retrieval to adapt as intermediate evidence reveals new information needs. Existing approaches support such retrieval through explicit graph structures or iterative query reformulation, introducing graph-construction or query-time overhead. We present Evidence-Guided Adaptive Graph Retrieval (EAGR), a framework in which accumulated evidence dynamically determines which semantic directions to explore next, and which requires no dedicated graph database and no explicit graph materialization. EAGR operates over an implicit semantic graph induced by compact passage metadata pre-ingested into a standard VectorDB. Starting from dense-retrieval seed passages, an LLM iteratively assesses accumulated evidence, identifies promising semantic directions, expands the retrieval frontier through metadata-based neighborhood lookup, and terminates once sufficient evidence has been collected. By precomputing semantic metadata, EAGR enables evidence-conditioned graph exploration without materializing an explicit graph, maintaining a dedicated GraphDB, or iteratively rewriting the query. Across three widely used multi-hop QA benchmarks, EAGR achieves the highest LLM-based answer accuracy among evaluated methods, outperforming the vanilla RAG by 17\% and strongest graph-aware baseline by 4\% percentage points averaged across three popular benchmarks HotPotQA, 2WikiMultiHopQA, and MuSiQue.

0.18

GSExplorer: Interactive World Generation with a Persistent 3D Gaussian World State

cs.CV cs.AI ·2026-11-01 ·Sicheng Yu, Liu Liu, Xinjie Wang, Xinjiang Wang, Taojun Ding +5 more
Abstract · 17 pages

Interactive world generation requires both accurate local control and stable scene consistency over long rollouts. Existing methods improve local generation through camera or geometry conditioning and extend historical context with memory mechanisms. However, long-horizon exploration remains challenging as target-view guidance becomes incomplete and inconsistencies accumulate across repeated scene updates. We present \textbf{GSExplorer}, a generation--reconstruction framework built around a persistent 3D Gaussian world state. At each step, the current 3DGS scene provides multimodal rendering conditions for a condition-driven video generator, while camera geometry is injected through a separate Relative Ray Encoding branch. Generated videos are then fed back to a SLAM-based reconstruction backend, where inconsistency filtering prevents conflicting content from corrupting the scene and cross-chunk bundle adjustment improves spatial alignment across generation steps. We further design a data-construction pipeline that reproduces realistic 3DGS degradation during scene extrapolation, yielding 293K training pairs of degraded multimodal renderings and real target videos. GSExplorer achieves the best overall performance among the compared methods on both single-step and three-step WorldScore evaluation, with strong camera control and consistency across both settings, including a Camera Control score of 99.07 in single-step extrapolation. On 481-frame leave-and-return rollouts, it also shows strong revisit consistency, demonstrating the value of maintaining a reliable 3D world state for continued exploration.

0.18

Restoring Semantic Fidelity for Training-Free High-Resolution Image Generation in Diffusion Transformers

cs.CV ·2026-11-01 ·Junjie Li, Zengbin Wang, Gaoxiang Li, Muyi Sun, Yixin Zhang +1 more
code/data
Abstract · accept by acm mm2026

Diffusion Transformer (DiT)-based text-to-image models have recently achieved remarkable success in image synthesis. However, when directly applied beyond their training resolution, these models often suffer from severe semantic degradation, where meaningful semantic content disappears. Although existing training-free high-resolution generation methods can mitigate this issue, the inherent semantic degradation at high resolutions remains largely unresolved, leading to degraded fine-grained details. Furthermore, widely used structural guidance methods lack modeling of the structure and details in the intermediate trajectory, which degrades the quality of the generated results. In this paper, we propose a training-free framework for high-resolution image generation to address these challenges. First, we propose a text-visual scale-decoupled attention (TVSDA) module, which restores the model's ability to generate semantically meaningful content at high resolutions by preserving the global receptive field of text tokens while aligning the receptive field of image tokens with the training resolution. Second, we propose a structure-detail guidance (SDG) strategy that explicitly models structural information and fine details through scale-wise and temporal-wise guidance, overcoming the limitations of conventional low-resolution guidance. Extensive experiments demonstrate that our method significantly improves both structural coherence and detail fidelity over existing training-free DiT-based high-resolution generation approaches. The code is available at https://github.com/ljjcoder/Restoring-Semantic-Fidelity.

0.17

Label-Free Localization and Test-Time Suppression of CLIP's Typographic Circuits

cs.CV ·2026-11-01 ·Bohan Liu, Wenqian Ye, Guangzhi Xiong, Zhenghao He, Sanchit Sinha +1 more
Abstract

CLIP-based Vision-Language Models have become cornerstones of modern multimodal systems. The cross-modal contrastive objective equips the CLIP vision encoder to both see objects and read text. These two abilities induce an entanglement between the encoding of lexical and semantic shapes, creating a distinctive vulnerability to typographic attacks, in which rendered text overrides an image's semantic content with its literal meaning. The black-box nature of these encoders has prevented a mechanistic interpretation of such entanglement. We introduce a head-decomposed Sparse Dictionary Learning (SDL) architecture that, by design, decomposes individual attention heads into interpretable, head-specific latents. Based on this inductive disentanglement, we re-purpose Clarity Score and propose normalized Text Attribution Score, which are both label-free, to identify attention heads that carry typographic information. As a causal validation of successful circuit separation, we perform a mechanistic intervention that reweighs attention away from the patch tokens in the identified lexical heads at test-time. Experiments show that the intervention effectively recovers object classification under typographic attack and, across five CLIP variants and four typographic benchmarks, achieves up to 46.9% improvement while incurring up to 1% accuracy loss in general image perception, outperforming prior supervised and training-free methods. Expanding on the mechanistic intervention, we propose a test-time adaptive attention head routing to further reduce the impact on performance at matched robustness, pushing beyond the Pareto frontier of current robustness-perception tradeoff.

0.17

MoPort: A View-Aware 2D Interface for Motion Language Models

cs.CV ·2026-11-01 ·Lixing Tan, Qing Xia, Yuting Guo, Shuai Li, Aimin Hao
Abstract

Motion language models (MoLMs) encode 3D human motion as discrete tokens for language-based understanding. Deploying them on monocular video raises two problems. Pose detection yields 2D joint sequences, leaving a dimensional gap from the 3D motion used in training. Moreover, the 2D observation changes with the camera viewpoint, while the depth lost in projection makes the 2D pose ambiguous across viewpoints. We propose MoPort, a view-aware 2D interface that enables MoLMs to take 2D observations as input. We introduce Paired View Alignment (PVA), which gives two views of the same motion a shared 3D teacher target and directly constrains the consistency of their latents, so that 2D observations from different viewpoints are mapped to a consistent semantic representation. Camera-Modulated Encoding (CME) estimates the camera extrinsics from the 2D sequence and uses them to modulate encoding, providing the 2D latents with geometric information about the viewpoint. MoPort achieves R@1 of 0.5674 on HumanML3D and 0.3286 on KIT-ML, outperforming all evaluated baselines. On HumanML3D, its R@1 decreases only from 0.5763 on in-distribution viewpoints to 0.5585 on out-of-distribution viewpoints. We validate portability on three downstream MoLMs. Furthermore, on the real-world RGB video dataset Fit3D, MoPort achieves 80.36\% accuracy. Code will be released upon acceptance.

0.17

Reconstruction as Spatial Evaluation: Diagnosing Cross-View Spatial Intelligence in Large Vision-Language Models

cs.CV cs.AI ·2026-11-01 ·Junfei Wu, Qiuyue Wang, Xuhong Huang, Qihang Peng, Mingsheng Li +6 more
benchmark
Abstract · 36 pages, 12 figures, including appendices

The spatial intelligence of large vision-language models (LVLMs) requires the ability to combine local visual observations into a geometrically consistent representation of the physical world. Many static benchmarks assess isolated spatial judgments, such as relative direction or distance, while interactive embodied evaluations combine perception with exploration, planning, control, and simulator-specific execution. To bridge this gap, we introduce RECONSPATIAL, a benchmark centered on multi-view 3D scene reconstruction across indoor, tabletop, and outdoor environments. Given real multi-view images and target-object queries, LVLMs are required to jointly predict oriented 3D bounding boxes in a shared, camera-anchored metric coordinate frame. Reconstructions are evaluated along two complementary axes: Structure for relative layout and Metric for physical scale and camera-anchored placement. Nine atomic spatial QA tasks, grouped into Geometric Perception and Spatial Relations, probe spatial capabilities involved in joint reconstruction. Offline navigation extends evaluation to path planning. Evaluation of 18 frontier LVLMs reveals a consistent gap between relative scene structure and metric reconstruction; even GPT-6-Astra scores 81.4% on Structure but only 58.7% on Metric. Tracing this gap, diagnostic analyses link atomic spatial capabilities to reconstruction quality and show that relational consistency alone does not ensure metric accuracy. Experiments with 3D map injection further separate scene construction from path planning, revealing how these geometric errors limit navigation success. Finally, spatial fine-tuning yields broad transfer across both spatial and general vision-language benchmarks.

0.17

Accounting for Irregular State Transitions through Belief Estimation in Online Medical Consultation

cs.CL ·2026-11-01 ·Mingzhe Lu, Xiangru Chen, Jian Qin, Yunpeng Li, Xihan Gao
Abstract · 12 pages, 2 figures, 3 tables

While online medical consultation broadens access, unprompted patient disclosures and topic shifts frequently disrupt clinician inquiry, leaving clinical states progressively revealed. We characterize these dynamics using both existing evaluation metrics and task-specific interaction metrics designed in this work. Through an empirical study across 16 medical dialogue datasets, we show that authentic online interactions exhibit substantially lower contextual continuity and higher unpredictability than offline and simulated benchmarks. We propose PRISM (Policy Reasoning with Inferred States for Medicine), a framework inspired by partially observable Markov decision processes that decouples latent state estimation from action selection to address drifting dialogue trajectories. Specifically, PRISM factorizes clinical triage into belief estimation, policy optimization, and response realization. By maintaining an evolving posterior belief over consultation states and conditioning action selection on the maximum a posteriori state, PRISM advances clinical goals while accommodating patient-initiated shifts. Extensive evaluations show that PRISM substantially outperforms frontier LLMs and clinical agents, achieving a 22.40 percentage point gain in clinical decision accuracy alongside decisive improvements in response generation, surpassing the strongest baseline by 5.51 points in BERTScore_F1. Furthermore, ablation studies show that providing oracle consultation states consistently boosts response quality across all evaluated baselines by guiding their action reasoning, yielding average gains of 5.58 points in BERTScore_F1.

0.17

dKV Forcing: Separating Generation and Memory Representations through a Differentiable KV Cache

cs.CV cs.AI ·2026-11-01 ·Zhiqi Li, Bo Zhu
Abstract

A causal video generator produces two key--value (KV) states for every segment: the KV it uses while denoising that segment, and the KV it writes into the cache for later segments to read. In a pretrained generator the two already differ substantially, later segments attend to them differently, and replacing the written KV with the generation KV degrades the continuation. We propose dKV Forcing (dKV denotes differentiable KV) to learn a dedicated memory representation from its future use: a writer-specific low-rank residual on the shared backbone separates memory writing from generation and is trained through the future reads of what it writes. Learning a write from its future use needs gradients through the cache, which the mutable cache in common use cannot provide. dKV Forcing replaces the training-time cache by persistent differentiable pages with local rematerialization and, in one temporal traversal, computes exactly the gradient that a two-pass reconstruction of the same objective computes. One traversal computes the full memory gradient at $7\%$ generator-update overhead over detached-write training, with $28\%$ less resident memory and $20.8\%$ shorter training cycles than the two-pass reconstruction. Over Self Gradient Forcing (SGF), the same objective trained with the same traversal, writer adaptation raises VBench-Long aesthetic quality by $6.6$ points at 60\,s and $8.5$ points at 240\,s with the highest subject and background consistency, and rollout dropout of the residual keeps most of this gain while restoring the motion of training without write gradients.

0.16

FOLLIE: Automated Reasoning over the First Dataset of Sentences with their First-Order Logical Representation for French

cs.CL ·2026-11-01 ·Maximos Skandalis
Abstract

FOLLIE (for First-Order Logic for Language Inference and Entailment for French) connects French sentences, compositional first-order representations and automated natural language inference (NLI). The first French NLI-ready, NL-to-FOL resource, it contains 19,983 inference records and 36,987 non-empty sentence-slot--formula mappings from six heterogeneous NLI sources, retaining French lexical material and source problem structure. Its silver annotations are produced by a neural--symbolic pipeline, with 111 separately identified LLM-assisted additions. The construction combines morphosyntactic analysis, probability-conditioned categorial multitagging, grammatical proof search, discourse representation and first-order semantic construction. Stage-wise experiments show substantial gains in derivation coverage; analyses of conjunction distributions, logical-operator incidence and scope alternation characterise the resulting resource. We then develop an automatic inference engine whose input-conditioned background theories address lexical relations, quantification, quantities, event structure and temporal dependencies. Prover9 supplies proofs for entailment and contradiction, while Mace4 supports bounded consistency checks. Three-way accuracies are 65.37% on all 335 FraCaS-FR problems, 70.33% on 300 GQNLI-FR problems and 71.35% on 9,399 eligible SICK-FR records pooled across splits. Agreement with reference labels among committed decisions is 93.53%, 100.00% and 98.66%, at commitment coverage of 41.49%, 37.00% and 15.05%, respectively. Worked constructions, axiom schemas and proof analyses explain these outcomes and identify the contributions of scope, reference, lexical knowledge and numerical structure. FOLLIE provides a French experimental basis for studying the representations and assumptions that connect sentence interpretation to deductive inference.

0.16

LUMEN: Geometry-Guided Diffusion Model for Spatially-Varying HDR Illumination Estimation

cs.CV ·2026-11-01 ·Hongze Chen, Fangyu Lin, Runze Zhang, Lingting Zhu, Weikai Chen +6 more
Abstract

Estimating scene illumination from images is challenging as lighting is inherently spatially varying, depending on geometry, occlusion, and light sources. Existing methods incorporate geometric cues by deriving local environment maps or probe-based representations, but largely rely on regression or parametric models, resulting in over-smoothed lighting and limited realism. We revisit spatially-varying illumination estimation from a generative perspective and introduce LUMEN, a geometry-conditioned diffusion framework for synthesizing high-dynamic-range (HDR) environment maps at arbitrary spatial locations from single or multi-view images. Specifically, we reconstruct a coarse 3D scene from input images and render partial environment maps at arbitrary locations to encode visibility and spatial structure. These signals are fused with multi-view image features and fed into a diffusion transformer to generate full HDR environment maps. To train the model, we build a synthetic dataset of 700 indoor scenes with dense spatial supervision, including multi-view RGB images, depth, and HDR illumination at multiple sampled locations. This dataset enables the model to learn spatially consistent lighting distributions across diverse scenes. Extensive experiments demonstrate that LUMEN significantly improves visual fidelity and spatial coherence compared to prior approaches.

0.14

LegoEnv: Self-Evolving Agentic Environments via Modular Recombination

cs.CL cs.AI ·2026-11-01 ·Zhuosong Jiang, Chengxu Wu, Junqi Liu, Yujie Yi, Siqin Wang +9 more
Abstract

LLM agents learn by interacting with environments, yet existing environments are designed for specific tasks and fail to generalize to new ones. Recent methods automate environment construction with LLMs. However, they generate each environment from scratch, neither reusing parts shared across environments nor carrying a fix from one environment to the rest. To address this, we propose a new paradigm of evolving environment atoms, which composes environments from reusable atoms (intent, rule, data and server) and evolves the atoms when an environment fails to give reliable feedback (e.g., an application lacks a tool the task requires). To instantiate this paradigm, we introduce LegoEnv, which builds 5,465 verified environments over 223 applications, and one round of evolution raises the share of solvable environments from 71.6% to 88.3%. On AutomationBench, SFT on these environments lifts Qwen3.6-27B from 16.0% to 32.0%, and RL with their verifiers lifts Qwen3.8-27B from 37.6% to 40.3%. Further analyses show steady gains with corpus size, transfer to unseen applications, and failures concentrated in a few shared root causes.

0.14

MUPA${}^2$S: Multimodal Unified Perception with Asymmetric Attention for Stress Assessment

cs.CV cs.AI ·2026-11-01 ·Stefanos Gkikas, Thomas Kassiotis, Yu Fang, Christian Arzate Cruz, Miao Cheng +6 more
Abstract

Stress is a leading contributor to long-term physical and mental health decline. The automatic assessment of stress is a promising approach for continuous monitoring and mitigating human biases. However, the development of automated systems remains challenging due to the heterogeneity of physiological and behavioral signals. This paper presents MUPA${}^2$S, a multimodal framework for automatic stress assessment that combines facial video and ECG signals within a unified perception architecture. Both modalities are mapped into a shared token representation without modality-specific components, and processed through asymmetric attention comprising segment-local cross-attention and global self-attention across segment states. Experiments on a newly collected 58-subject stress dataset, evaluated under a stratified subject-level protocol, show that facial video achieves 69.44% test accuracy, while ECG-only performance remains close to chance under the proposed representation. Unified Video-ECG fusion achieves a test accuracy of 69.58% at approximately half the number of parameters of feature-level fusion, with negligible additional computational cost. Feature-level fusion does not exceed the video-only baseline, while unified input-level integration provides a parameter-efficient mechanism for jointly processing the two modalities.

0.14

Depth Anyviewpoint? Evaluating Monocular Depth Over Viewpoint Sequences

cs.CV ·2026-11-01 ·Theo Rode, Rui Zhang, Ryan Aubrey, Olivia Li, Jessica Huang +3 more
Abstract

Monocular depth estimation aims to predict depth maps from single RGB images (forgoing the need for stereo camera pairs) and has seen impressive leaps in performance in recent work due to the introduction of affine-invariant losses and the use of foundation model features. These monocular depth models predict affine-invariant depth, which describes the relative positions of objects in the scene instead of absolute depth. While these models demonstrate strong performance on single image benchmarks and qualitatively consistent predictions in video, their effectiveness at preserving consistent spatial structure has been less thoroughly explored. Many potential real-world applications, such as handling video streams as input, rely on coherent three dimensional consistency in scenes. Further, anchoring relative depth predictions to a metric reference for a given scene is a key challenge for applying monocular depth models to real-world situations. We develop a novel evaluation protocol for monocular depth models using video sequences with moving camera positions to test the consistency of prediction across variation in viewpoint. We demonstrate this methodology over diverse, real-world video sequences where we are able to anchor depth model output to metric depth values using structure from motion. Our work demonstrates open challenges in the development of broadly applicable monocular depth foundation models, particularly with their ability to provide temporal consistency over long time frames. We also identify challenges in the straightforward adoption of structure-from-motion as a metric grounding tool for monocular depth estimation models.

0.14

OCBI: Order-Sensitive Cross-Branch Interaction for Image Manipulation Localization

cs.CV ·2026-11-01 ·Zhouli Yu
Abstract · 13 pages, 3 figures, 4 tables

Image manipulation localization combines semantic structure with forensic traces to identify modified pixels. Within-branch spatial mixing and RGB-guided cross-branch mixing can produce different messages when composed in opposite orders. We propose OCBI, an extension of Co-Transformers that models the discrepancy between these two composition orders for localization. A local cross-branch fusion module uses the commutator of RGB-guided cross-attention and forensic self-attention to add a learnable signed correction to feature transfer. Pixelwise adaptive multiscale aggregation then combines spatial contexts before prediction. OCBI uses the Co-Transformers backbone and supervision. We compare OCBI with retrained Co-Transformers using CASIA-only task training, a 50-epoch budget, and final-epoch evaluation. Across five clean test datasets, OCBI achieves a macro PixelF1 of 0.5773, exceeding the baseline by 0.43 percentage points. Under JPEG compression, Gaussian blur, and resizing on three datasets, its average PixelF1 over 21 dataset-transformation conditions exceeds the baseline by 0.95 percentage points.

0.14

FinishStrong: Learning to End Well after Imperfect Start in Video Temporal Grounding

cs.CV ·2026-11-01 ·Yuda Zou, Boxiang Zhou, Xin Zhou, Yibo Chen, Dejia Song +3 more
Abstract

A poor start lowers the ceiling, but does not negate the value of finishing well. In autoregressive video temporal grounding, models systematically exhibit an endpoint performance asymmetry, predicting ends significantly worse than starts. A committed flawed start acts as an immutable physical constraint, irreversibly capping the attainable overlap as a sunk loss. While models should minimize the remaining conditional regret, standard post-training paradigms fail to teach this adaptation. Specifically, teacher-forced SFT suffers from strict prefix exposure bias, and holistic trajectory evaluation in standard GRPO inadvertently penalizes conditionally optimal ends if they follow a poor start. To address this, we introduce FinishStrong, a novel two-stage post-training framework explicitly designed to teach optimal continuations. First, Online-start SFT exposes the end prediction to the model's own generated contexts. Second, Conditional-tree GRPO restructures the sampling topology, centering end rewards exclusively within the same-start branch to mathematically isolate the conditional end quality from the inherited start penalty. FinishStrong-9B achieves a state-of-the-art 61.7\% overall IoU on TimeLens-Bench, remarkably outperforming flagship closed-source models like Gemini-3.8-Flash and Qwen3.8-Max. Crucially, when forced to continue from imperfect starts, it reduces the subsequent end prediction error by 36.5\%, proving its exceptional ability to end well after flawed starts. Code will be open.

0.14

Towards Self-Evolving Narrative Systems: Real-Time Adaptation and Long-Term Optimization with LLMs

cs.CL ·2026-11-01 ·Ruisheng Sun, Jiaqi Shao, Weijia Han, Linxi Ju, Yunzhen Zhang +7 more
Abstract · Under review at the ACL Rolling Review (ARR) October 2026 cycle

Traditional interactive narrative systems offer strong authorial control and structural coherence but often restrict player agency. In contrast, LLM-driven narratives enable open-ended generation and greater flexibility, yet frequently suffer from inconsistency and diminished narrative control. To navigate this trade-off, we propose a hybrid narrative framework that dynamically balances pre-authored content with real-time LLM generation via semantic similarity alignment. Additionally, sustaining narrative quality over extended gameplay remains a core challenge in LLM-based storytelling. To address this, we introduce a self-evolving, feedback-driven optimization framework that leverages in-context learning to iteratively refine prompts, eliminating the need for manual prompt engineering. Through a case study and participant-based evaluation, we show that our framework enhances narrative coherence while preserving player agency, effectively aligning generated content with both authorial intent and evolving player behavior.

0.14

Revisiting LLM Homogeneity Through a Multidimensional Lens

cs.CL cs.AI ·2026-11-01 ·Ziliang Qiu, Renfen Hu, Yi Ding, Ted Underwood
Abstract · 26 pages, 9 figures

Concerns about LLM homogeneity are typically quantified with embedding similarity, which reduces two texts to a single score. We revisit homogeneity with 25 interpretable metrics spanning five dimensions: semantic, lexical, syntactic, structural, and stylistic & presentational. A human study shows that perceived similarity depends on both content (what responses say) and form (how they say it), yet embedding similarity correlates less with human meaning judgments than sentence-level matching ($ρ=0.417$ vs. $0.606$) and not significantly with form judgments ($ρ=0.140$). Across 18 contemporary LLMs, models say similar things in different words: they share global semantics and broad linguistic profiles but remain distinct in lexical characteristics, structural patterns, and interactional conventions. Across releases from 2024 to 2026, models become more like themselves but not more like one another: among significantly changing metrics, intra-model distances decrease for 17 of 22, and inter-model distances increase for 14 of 21. Across two post-training pipelines, model-specific conventions emerge during SFT and are selectively reshaped by preference optimization. These results suggest that homogeneity is multidimensional and evolving, and that embedding similarity alone can overstate cross-model homogeneity while concealing diverging intra- and inter-model trends.

0.13

GaitFM: A Cross-Modal Foundation Model for Wearable Gait Assessment

cs.CV ·2026-11-01 ·Yize Cai, Zhiqing Hong
Abstract · 18 pages, 4 figures

Gait assessment supports the diagnosis, monitoring, and prognosis of diverse movement-related conditions, such as neurological and musculoskeletal disorders. Recent advances in deep learning using everyday wearable inertial measurement units (IMUs) offer an accessible and low-cost approach to automated gait assessment. However, IMU measurements vary substantially across individuals, body locations, and device configurations, while existing deep learning models are typically developed using small, homogeneous datasets for specific clinical tasks, limiting their robustness and generalizability in diverse real-world contexts. Here we present GaitFM, a cross-modal foundation model for gait assessment using IMU sensors. GaitFM improves the generalizability of gait assessment by transferring knowledge from large vision models and pretraining on unlabeled IMU sensor data. We pretrain GaitFM on a large curated dataset comprising about 4,000 participants and 1.23 million gait windows across diverse clinical and daily conditions, sensor modalities, and body locations. GaitFM demonstrates strong performance across 18 gait-related tasks, including 3 gait parameter estimation tasks, 14 disease and gait abnormality detection tasks, and 1 clinical outcome prediction task. Furthermore, GaitFM generalizes across different sensor configurations and body locations, including those commonly used in clinical settings and everyday wearable devices. These findings demonstrate the effectiveness of transferring visual knowledge in improving sensor-based gait assessment performance, offering a new promising pathway for addressing data scarcity in the sensor and medical domains.

0.13

Recoverability-Aware Evaluation of Diffusion Video Inpainting and Object Removal

cs.CV ·2026-11-01 ·Yanliang Qi, Jiacheng Ding, Cong Guo, Peihua Mai
Abstract · 10 pages, 6 figures, 7 tables

Video inpainting for object removal is scored over the whole hole, although the background behind the object is visible in other frames of the shot for some hole pixels and in no frame for others. A whole-hole score therefore mixes copying with generation and cannot show which of the two a remover does well. To separate the two, we propose recoverability-stratified evaluation, in which a point tracker run on the clean video labels each hole pixel by whether and how far away it is observed, so that fidelity and temporal stability can be reported per stratum against a copy reference. To reuse the observed content, we also propose evidence-guided sampling (EGS), which blends tracked colors into the clean-latent prediction of a video diffusion remover without training. On DAVIS, the perceptual advantage of three diffusion removers over flow propagation shrinks where the content is visible a few frames away (LPIPS difference-in-differences +0.046 to +0.056 against E2FGVI-HQ), and real paired captures show the same pattern for two of them. Their DAVIS hole warping error is two to four times that of ProPainter, and copying along clean-video tracks lowers their recoverable LPIPS by a quarter to two fifths. EGS closes 43% and 56% of the gap to this copy reference on held-out DAVIS when its colors come from clean-video tracks. With colors tracked on the masked input, a pre-registered confirmation finds no gain on MiniMax-Remover and a loss on ROSE. Diffusion removers thus leave observed content unused on DAVIS, and training-free reuse does not yet hold with tracked evidence.

0.13

QCVS: Question-Contrastive Visual Steering for Multimodal Reasoning

cs.CV ·2026-11-01 ·Tong Fu
Abstract · 5 pages, 2 figures, 3 tables

Inference-time steering in vision-language models often entangles query-relevant reasoning dynamics with low-level perceptual noise, causing collateral degradation on correct predictions. We propose Question-Contrastive Visual Steering (QCVS) to isolate task-specific intervention directions via double-differential activation probing. By contrasting spatial layout perturbations under the target query against a generic description reference, QCVS cancels task-agnostic visual artifacts without ground-truth supervision. A dual-threshold gating mechanism then evaluates relative interaction strength and signal amplitude, selectively applying a single bounded hidden-state update only when necessary. Empirical evaluations across multiple scales of Qwen3.5 (2B to 9B) and DeepSeek-VL2 on DynaMath and We-Math demonstrate that QCVS achieves state-of-the-art accuracy gains under sparse coverage (approximately 22%-30%). Crucially, QCVS yields repair rates up to 13.98% while suppressing collateral damage down to 0.41%, proving high intervention precision and computational efficiency.

0.13

Tracing How Language Models Integrate Temporal Anchors into Factual Recall

cs.CL ·2026-11-01 ·Ayana Niwa, Masahiro Kaneko, Kentaro Inui
Abstract

Factual recall in language models is typically framed as retrieving a factual association: a subject--relation pair maps to a single object. However, many facts are context-dependent: the same pair maps to different objects depending on a condition such as time. We introduce anchored factual recall, in which the model receives an explicit conditioning expression, an anchor, and must select the object valid under it. This raises questions that single-object recall does not: where the anchor enters the computation, what it changes, and how that change selects the answer. Focusing on temporal anchors, we answer them with causal interventions on four open-weight models on English prompts and identify two routes. In a subject-mediated route, the anchor resolves the final subject-token representation to the queried year, and the relation redistributes the readout of this state across attention heads. In a direct route] the final position attends to the anchor, and upper-layer MLPs turn this signal into a preference for the time-appropriate object. The two routes are parallel channels: their contributions add up, most queries use both, and their balance depends on the query, e.g., on the relation, consistently across models. This reframes mechanistic accounts of factual recall from retrieving a single stored object to selecting among context-dependent alternatives.

0.12

GDBack: A Group-Delay Backdoor Attack on Speaker Identification via Analytic All-Pass Filtering

cs.SD eess.AS ·2026-11-01 ·Bicheng Xiong, Tao Li, Chenhui Cui, Zedong Xing, Qien Li +3 more
Abstract

Existing phase-based backdoor attacks on speaker identification rely on shared absolute phase patterns for cross-sample trigger consistency, while other audio backdoor attacks have explored shared transformations that produce input-dependent triggers. Whether phase-based triggers can achieve similar consistency without imposing identical output phase patterns remains unclear. We propose GDBack, a poison-only group-delay backdoor attack using cascaded second-order analytic all-pass filters. Applying a shared transformation across utterances ideally preserves spectral magnitude while introducing a consistent frequency-dependent phase response and group-delay contribution, without forcing identical output phases. Experiments on LibriSpeech and VoxCeleb1 with ECAPA-TDNN, x-vector, and RawNet3 achieve attack success rates above 94% across all six settings (97.00% on average), while maintaining benign accuracy close to clean models. Phase and magnitude interventions, inverse compensation, and parameter-sensitivity analyses support the role of the induced phase transformation in backdoor activation. Speech quality and content evaluations indicate limited perceptual and linguistic distortion, while robustness experiments, including real over-the-air tests, reveal varying effectiveness across conditions and architectures. These findings demonstrate that cross-sample trigger consistency can arise from a shared phase transformation without requiring a common absolute phase pattern.

0.12

Knowledge-Guided Floodwater Level Reasoning for Cloud-Edge Collaborative Monitoring

cs.CV cs.AI ·2026-11-01 ·Zhengshu Zhou, Shaoyu Wang, Ziyi Geng, Lu Tao, Qian Long
benchmark
Abstract

Frequent urban floodwater poses a serious threat to the safety and efficiency of intelligent transportation systems. Existing vision-based methods for floodwater level detection face two main limitations. First, the lack of explicit reasoning leads to poor detection performance and little interpretability. Second, these methods struggle to generalize to the complex and varied real-world road scenes seen from different viewpoints. To address this, we reformulate the task as an explicit visual reasoning problem and propose HydroReasoner, a floodwater level reasoning framework designed for cloud-edge collaboration. HydroReasoner builds on Qwen2.5-VL and is trained with Group Relative Policy Optimization. To address the scarcity of reasoning process annotations, we build a floodwater domain knowledge graph HydroKG and design a novel KG path reward. Through deterministic graph traversal over HydroKG, this reward supervises the model's reasoning process at a fine granularity, which improves both reasoning and generalization. For edge deployment, we apply 4-bit quantization to the model, cutting GPU memory usage by 57% and inference latency by 43% while out-of-domain accuracy drops by only 0.2 percentage points. We also propose a confidence-based cloud-edge routing mechanism, which improves system safety and eases the computational load on the cloud. In addition, we build HydroLevel, a floodwater benchmark dataset of 3,000 images, for training and evaluating the model. Experimental results show that HydroReasoner achieves detection accuracy 4.0 and 13.6 percentage points higher than conventional vision baselines in the in-domain and out-of-domain evaluations respectively, with performance dropping by only 4.4 percentage points out of domain.

0.11

NatureFusion: Rethinking Infrared and Visible Image Fusion as Natural Complementary Enhancement

cs.CV ·2026-11-01 ·Jianfeng Ding, Linfeng Tang, Hao Zhang, Han Xu, Xunpeng Yi +4 more
Abstract · 17 pages, 13 figures, 12 tables

In the absence of ground truth, most existing infrared and visible image fusion methods pursue fidelity to the source images and directly superimpose two physically distinct modalities. As a result, salient infrared targets appear as bright patches of thermal radiation pasted onto the scene. These patches obscure the original color and texture of the targets, so the fused images look unnatural to human observers and deviate from the data distribution learned by vision models pretrained on visible images. Therefore, we redefine infrared and visible image fusion as natural complementary enhancement rather than source fidelity, and propose NatureFusion, which exploits cross-modal complementary cues to infer a clear and natural visible-style image from low-quality visible and infrared image pairs. Specifically, to address the lack of ground truth, we employ a generative foundation model as the teacher to directly generate visually natural fused images for supervision. Furthermore, to fully distill the teacher's generative capability while reducing computational cost, we fine-tune a pretrained generative model with conditional flow matching through a light-weight fusion adapter. Last but not least, to further improve downstream performance without sacrificing natural appearance, we freeze the fusion adapter and introduce a detector-supervised task adapter that strengthens the complementary contribution of infrared. Extensive comparisons with seven multi-modal fusion and seven single-modality enhancement methods demonstrate that NatureFusion generates visually natural fused images with the best perceptual quality. Detection and segmentation models pretrained on visible images achieve substantially higher accuracy on our fused images than on those of any competing method.

0.10

Deep Learning for Cervical Cytology Screening: Methods, Algorithms, Challenges and Prospects

cs.CV ·2026-11-01 ·Jiang Yangbo, Zheng Nenggan
benchmark
Abstract

Cervical cancer remains a leading cause of cancer death among women, with more than 90\% of deaths occurring in low- and middle-income countries, where cytology screening is constrained by a shortage of trained cytopathologists. Deep learning has made automated cytology screening increasingly feasible, but the literature is fragmented across cell, patch and whole-slide-image scales, task formulations and supervision regimes, hindering comparison and clinical translation. This Review synthesizes the deep learning literature for cervical cytology screening, organizing the field along three axes---data scale, functional task and supervision---and distinguishing two clinical roadmaps: whole-slide-image confidence aggregation and cell-level detection with Bethesda classification. We examine 16 publicly available or publicly documented cytology datasets alongside the CAISHI histopathology dataset, compare quantitative performance while exposing incomparable metrics and data leakage, and assess commercial systems, regulatory status, HPV-era screening pathways and low-resource deployment. Single-cell accuracy on Herlev and SIPaKMeD has plateaued, but this reflects benchmark saturation rather than clinical readiness; prospective and randomized evidence is emerging but remains limited. We consolidate the evidence into five challenge areas---data and benchmarks, architecture and label-efficient learning, trustworthy AI, clinical translation and health economics, and evaluation and reproducibility---and provide an evidence-to-agenda map and a deployment decision matrix. Progress now depends on multi-center benchmarks, externally validated label-efficient methods, calibrated and fair predictions, and pragmatic clinical trials rather than further gains on saturated single-cell benchmarks.

0.09

RSI-Env: An Open-Ended Environment for Recursive Self-Improvement of Agents

cs.CV ·2026-11-01 ·Wanying Qu, Zhiyuan Fan, Jiyao Liu, Jiang Zhou, Jianghan Shen +2 more
benchmark
Abstract

Recursive self-improvement (RSI) envisions agent systems that autonomously improve their own models and harnesses, with each updated system continuing to improve itself. Existing approaches typically restrict optimization objectives, editable components, or evaluation procedures, limiting open-ended exploration and continuity across generations. We introduce RSI-Env, a persistent environment for coordinated model-harness improvement, supporting autonomous training, harness modification, and retention of research state. An independent Host Controller manages services, evaluation, and versioning. Under Self, each accepted model-harness version takes over the next improvement round. We define three settings that vary goal specificity and access to benchmark feedback: Open-General, Capability-Directed, and Benchmark-Guided. Across these three settings, we compare Self with two fixed external Teachers on seven tasks, starting from Qwen3.8-27B with a minimal harness. Without benchmark feedback, Open-General Self raises GPQA Diamond accuracy from 65.2% to 73.2% and AIME 2026 accuracy from 53.3% to 56.7%. All three Capability-Directed Self runs achieve target-aligned gains. Knowledge Self demonstrates continued improvement across system updates, with SimpleQA accuracy rising from 82.0% to 85.0% and then 88.0%. Benchmark guidance and Teacher benefits vary by task. These results demonstrate RSI-Env's support for open-ended exploration and sustained recursive improvement within the studied runs, while highlighting challenges in preserving gains across capabilities.

0.08

NAS-PO: Native Attention-Strategy Policy Optimization for Vision-Language Models

cs.CV ·2026-11-01 ·Ke An, Shizhe Xiang, Yue Liu, Liang Chen, Pei Fu +3 more
Abstract

Reinforcement learning with verifiable rewards (RLVR) has become an effective post-training paradigm for vision-language models (VLMs). Standard RLVR, however, assigns identical outcome-derived credit to all generated tokens regardless of their perceptual relevance. Recent perception-aware methods estimate visual reliance through image perturbation or masking. These indirect proxies require perturbation branches, while corruption sensitivity does not necessarily identify the visual evidence used for reasoning. We instead shape model perception through native attention, an intrinsic model signal. Rather than treating attention as a post-hoc explanation, we formulate it as a reward-anchored, token-aligned routing signal that encourages attention patterns supporting visually grounded reasoning. To this end, we propose NAS-PO, a Native Attention-Strategy Policy Optimization framework. NAS-PO analyzes native attention with a visual mass policy that quantifies visual allocation and a conditional visual policy that captures spatial routing over visual tokens, making the optimization well-posed. Based on this, Native Attention-Strategy Advantage Shaping (NAS-AS) draws on the visual-mass policy to redistribute signed policy-gradient credit for visually engaged rollouts. The Adaptive Attention-Coupled (AAC) loss then aligns the visual-mass policy on all response tokens and the conditional visual policy on visually active ones, consolidating the visual-routing patterns that drive successful trajectories. Experiments demonstrate that NAS-PO substantially surpasses leading RL-tuned VLMs across 7 out-of-domain reasoning and perception benchmarks, with its effectiveness consistently validated across 8B and 32B. Visual-Routing Sensitivity (VRS) analysis further indicates that the optimized attention prioritizes visually load-bearing regions.

0.07

4D Triangle Splatting Reconstruction of Dynamic Endoscopic Scenes from Monocular Videos

cs.CV cs.RO ·2026-11-01 ·Laura Salort-Benejam, Stamatia Giannarou, Antonio Agudo
Abstract · Off-Grid; MICCAIW 2026

Robust 4D reconstruction of deformable endoscopic scenes from monocular videos remains challenging due to non-rigid tissue deformation, illumination changes, occlusions, and unknown camera trajectories. Reliable dynamic reconstruction of endoscopies could significantly enhance visualization, diagnostic accuracy and intraoperative guidance of minimally invasive procedures. We propose SurgicalTS, a self-supervised triangle-splatting framework for dynamic endoscopic scene reconstruction from static monocular cameras. The scene is represented by a canonical set of triangles jointly optimized with a deformation network. We introduce a quaternion-based deformation model, an improved color parametrization, spatio-temporal priors, and a redefined initialization strategy that improves completeness under occlusions. Extensive evaluation on challenging in-vivo endoscopy videos from diverse procedures shows that SurgicalTS successfully models soft-tissue deformations and tool-tissue interactions over time, while consistently outperforming state-of-the-art methods. Our method also reduces training times while maintaining competitive geometric and photometric accuracy results, highlighting the effectiveness of quaternion-based deformation modelling with physical consistency for robust and efficient 4D reconstruction of endoscopic scenes.

0.06

Fréchet Oral Distance: A Coordinate-Free Metric and Benchmark for Speech-Driven Oral Animation

cs.CV cs.GR ·2026-11-01 ·Jaehong Myung, Seungho Eum, Hyunjung Chung, Unsang Park
benchmark
Abstract

Recently, speech-driven facial and oral animation technologies have been used in various fields such as education, speech therapy, and healthcare. But methods to objectively evaluate the quality of generated results are still limited. Accordingly, we propose the Fréchet Oral Distance (FOD), which uses embeddings from an Oral Motion Encoder (OME) and a TIMIT-Vis reference distribution. In experiments, FOD reproduced model rankings similar to the coordinate-based metric, Lip Vertex Error (LVE), and showed trends consistent with user preference. Our framework provides objective and reproducible benchmarking without relying on expensive equipment or subjective evaluation, and it can be extended to compare a variety of 2D/3D models.

0.06

Security & Privacy 17

CipherFlow: Compiling AI Programs for Efficient Cryptographically Secure Inference

cs.CR cs.LG ·2026-11-01 ·Hedong Zhang, Mengxin Zheng, Qian Lou
top venue
Abstract · Accepted at NeurIPS 2026. 34 pages. Code and interactive demo: https://loopleap-lab.github.io/CipherFlow/

Cryptographically secure inference faces two practical challenges: making secure programming accessible to AI developers and data scientists, and achieving efficient execution across deployment environments. Developers must compose cryptographic primitives correctly, while performance depends on hardware, networks, and evolving ciphertext states. We present CipherFlow, a compiler framework that takes a high-level AI computation graph, available cryptographic primitives, and a target hardware and network configuration to generate an executable, latency-optimized secure inference program. CipherFlow automates cryptographic execution decisions through three techniques: MicroBench profiles deployment-specific primitive costs; state-expanded planning jointly optimizes homomorphic encryption (HE) and secure multi-party computation (MPC) placement, conversions, and ciphertext maintenance under cryptographic-state constraints; and BlockFold summarizes branched regions to scale planning to Transformer graphs. Evaluation using measured HE/MPC backend costs across CPU/GPU and network settings shows up to 16.5$\times$ speedup on BERT-base and over 22$\times$ on ViT compared with static hybrid placement baselines. CipherFlow separates model programming from cryptographic execution decisions, enabling developers to express AI computations while the compiler constructs efficient secure programs.

0.62

Increasing the security of advanced metering infrastructure using dynamic defense methods

cs.CR ·2026-11-01 ·Mohsen Saffar, Hamid Reza Naji
published
Abstract

With the increase in Internet usage over the last half century, technology has also benefited from the growth and development of the Internet. It is no wonder that smart grids are emerging as one of the most popular technologies of today. As smart grids are connected to the internet, they are constantly vulnerable to cyberattacks. It is always possible for attackers to gain access to metering system infrastructure which is a major and critical part of smart grids. When this system is not working properly because of some security threats, irreparable damage is in the place for a country and its industries. This article suggests utilizing dynamic Applications that Participate in Their Own Defense defenses to secure such systems and infrastructures.

0.44

Detecting Generated-Depth Injection Attacks on RGB-D Place Authentication

cs.CR cs.CV ·2026-11-01 ·Seigo Kawachi, Satoshi Matsuura
published
Abstract · 6 pages, 1 figure, 1 table. Accepted to IEEE WIFS 2026. (c) 2026 IEEE

Advances in monocular depth estimation (MDE) now let an attacker generate depth from publicly available RGB images and submit forged RGB-D queries. Such attacks break a core assumption of RGB-D place authentication: that depth comes from a sensor. We study a presentation attack detection (PAD) classifier that uses geometric consistency between the enrollment and query views, as a software-layer defense. A three-stage evaluation shows how far this classifier detects software-injected depth, depending on how accurately the attacker knows the camera intrinsics. In the first stage, the attacker has only an estimate of the intrinsics, and the PAD classifier separates forgeries generated by all three MDE models with a validation area under the receiver operating characteristic curve (AUC) of 0.997-0.999. In the second stage, the attacker obtains the exact intrinsics, and the original PAD classifier falls to near chance (AUC 0.54-0.61); a retrained PAD classifier, however, still has the registration residual between enrollment and query as a cue. In the third stage, an oracle evaluation, in which the evaluator fits warps using the ground-truth 3D geometry, shows that detection falls to near chance only in its strongest case, where every valid query keypoint is given its ground-truth 3D coordinate. Of the detection cues, the reprojection error captures the attacker's misestimate of the intrinsics, whereas the registration residual remains even when the attacker knows the exact intrinsics.

0.39

Attack Patterns in Cross-chain Bridges: A Trust-based Empirical Analysis

cs.CR ·2026-11-01 ·Ishraq Samiha, Yasir Rahman, Rahma Mukta
code/data
Abstract · 18 pages, 15 figures. The associated dataset is available at https://doi.org/10.5281/zenodo.23271026

The rapid growth of multiple blockchain ecosystems has made cross-chain bridges essential for transferring assets and information across independent blockchains. Their multi-component architectures, cross-chain trust assumptions, and concentration of valuable assets expose them to diverse security threats. This paper presents an empirical study of cross-chain bridge attacks based on 48 evidence-supported real-world incidents from January 2022 to August 2026. We propose a behaviour-based attack pattern architecture comprising six trust-violation attack patterns, each defined by the primary trust assumption violated by the attacker: Proof Forgery,Logic Exploitation, Event Injection, Economic-State Manipulation, Credential Seizure, and Infrastructure Hijacking. The architecture further maps these patterns to five functional domains of cross-chain bridges: Deposit, Relay, Verification, Execution, and Governance and Operations, separating how a trust assumption is violated from where the compromised trust boundary lies. The 48 incidents correspond to $2.676 billion in gross bridge-specific losses, with the ten largest incidents accounting for 92.4% of total losses. Measured on the primary loss basis, Proof Forgery and Credential Seizure make up 86.2% of all losses, and the Verification and Governance and Operations domains hold 93.9%. A handful of very large incidents drives this concentration, yet it still holds when we switch to other loss bases or reclassify the boundary cases. By tying what attackers do to the bridge functions and security assumptions they target, the architecture gives threat modelling for cross-chain bridges an empirical footing.

0.36

Bounded Retained History in a Proxy Flow Replay: A Mixed Predictive Pilot and an Allocation Study

cs.CR cs.LGcs.NI ·2026-11-01 ·Juan S. Santillana
Abstract · 15 pages, 0 figures, 6 tables

We examine bounded retained history through a frozen predictive pilot and a subsequent allocation study on a curated UNSW-NB15 flow mirror. Both use an artificial release-order replay based on raw ltime ticks, with equal-tick records querying a shared prior-group state. In the pilot, the model with added retained relational features had lower reconnaissance recall than the current-flow-plus-counters baseline (43.66% versus 47.84%) and lower benign false-positive rate (3.63% versus 6.68%) at separately frozen operating thresholds, together with lower AUROC. This mixed result did not establish a predictive advantage. A development-only audit subsequently found that capacity eviction dominated expiration in the 256-record, 300-tick FIFO state, with a median completed record lifetime of 11 ticks. A protocol specified before its single execution, but after the pilot and post-hoc mechanism audit, compared five label-blind retention arms on the already-used development trace. Its primary equal-source endpoint averaged retained/reference ratios for service-expanded source out-degree and source-destination service diversity. Source-balanced retention at 256 records exceeded FIFO-256 by a pooled endpoint difference of 0.440738 and met the prespecified materiality and subblock rule. Crucially, the event-weighted ratios for both quantities decreased relative to FIFO-256: the result is specific to the equal-source weighting objective and does not establish uniformly improved retention. The study supplies neither evidence of detection improvement nor an explanation of the pilot result. Reused data, concentrated reconnaissance support, proxy timing, and the absence of a qualified independent corpus bound the interpretation.

0.33

Accelerating QUIC Encryption in WebAssembly Using WASM-SIMD

cs.CR cs.NI ·2026-11-01 ·Sota Yamawaki, Yasuo Okabe
Abstract · A Japanese version of this work has been submitted as a manuscript for presentation at the IEICE Technical Committee on Internet Architectur…

WebAssembly (Wasm) runs untrusted code in a portable, lightweight sandbox and is increasingly used to host services on shared servers and edge platforms. Running the QUIC transport protocol inside Wasm would let each application choose and update its own transport without changing the host. QUIC, however, encrypts every packet, and Wasm offers neither dedicated AES instructions nor carry-less multiplication; a Wasm QUIC stack has been reported to reach only 17.6% of native goodput, with cryptography as its largest cost. Whether the 128-bit SIMD extension of Wasm (WASM-SIMD) can recover this cost, and what limits a pure-Wasm stack once it does, has remained unclear. Here we show that explicit WASM-SIMD kernels recover much of the cryptographic cost, after which non-cryptographic costs become the limit. On two hosts connected by 10-Gigabit Ethernet, kernels that process eight AES blocks or four ChaCha20 blocks per pass raise receiver goodput over a non-SIMD baseline by 61.34% for AES-128-GCM and 31.55% for ChaCha20-Poly1305, whereas compiler auto-vectorization gains only 1.28% for AES-128-GCM. ChaCha20-Poly1305 still delivers 1.99 times the goodput of AES-128-GCM, confirming end to end the preference that stand-alone cipher benchmarks suggest. Receiver-side profiling combined with Amdahl's law predicts that even native-speed cryptography would leave Wasm at roughly one third of the goodput of the fastest native stack. Cryptographic acceleration is therefore necessary but not sufficient: memory allocation, runtime overhead and per-datagram UDP I/O now bound pure-Wasm transports. We anticipate that this per-category cost breakdown will guide further optimization of Wasm network stacks and the design of system interfaces for batched I/O.

0.33

Step-Zero Attack: Agent Safety Before Observability

cs.CR cs.AIcs.SE ·2026-11-01 ·Tian Qin, Yuhang Jiang, Junzhe Chen, Lit Sin Tan, Lijie Wen
Abstract · 14 pages, 5 figures, 4 tables

Existing attacks and defenses for coding agents act once content enters the model context: injected text persuades the model, and guards inspect what the model reads. Yet to judge whether processing an unknown archive, network response, or image is safe, an agent often must process it first, and that step can do harm before any content reaches the model. We call this first processing Step Zero, and harm done by it a Step-Zero attack. Across seven harnesses and six models, each agent received only a path or URL to a crafted input, paired with a benign twin showing the same visible fields. Step Zero is where most agents are already unsafe: 32/42 harness-model pairs overran a 4 MiB disk budget, 26/42 before the model saw any member, and four of the six models did so in every round (84/84), so the damage precedes any model-side defense. Every twin stayed within budget, so the attacker's bytes alone cause the harm. On network and image inputs the risk moves to what the agent writes: agents stored the body under the name the input declared, letting the input decide what is written, and their checks faded once cues were removed. Model-free tests trace the harm to the processing primitive, so we defend it there: a gate that counts the bytes each call produces held every routed extraction within budget while agents completed 25/28 tasks. Vendors should pair such gates with safety-alignment training that teaches agents to inspect an input before processing it.

0.32

ZT-DAC: ML-Driven Dynamic Access Control Policy for Zero Trust in DER Networks

cs.CR cs.AI ·2026-11-01 ·Arun B. Chandrasekaran, Damilola Popoola, Souradeep Bhattacharya, Manimaran Govindarasu
hot direction
Abstract · IEEE SmartGridComm 2026, College Station, Texas

The increasing penetration of Distributed Energy Resources (DERs) in the smart grid has significantly expanded the cyber-attack surface. Traditionally, DER networks rely on static role-based access control (RBAC) mechanisms that assume persistent trust after authentication. This limitation enables adversaries to maintain unauthorized privileges and move laterally across DER systems, disrupting grid operations. Although attribute-based access control (ABAC) enables fine-grained and context-aware access decisions, it does not fully eliminate implicit trust. Zero Trust (ZT) principles can be incorporated to enhance ABAC policies. This paper proposes a dynamic access control framework for implementing ZT in DER networks (ZT-DAC). The proposed solution employs a two-stage verification process and integrates an Open Policy Agent (OPA)-based RBAC with a machine learning (ML)-based ABAC to enforce continuous and adaptive trust score assessment. To support realistic DER operations, a DER-specific access control model is developed and used to generate a domain-informed dataset of access requests incorporating both legitimate and adversarial behaviors. Experimental results on the dataset show that the OPA-based decision layer achieves \(91.92\%\) accuracy, with a False Positive Rate of \(14.45\%\), while the ML model achieves a Mean Absolute Error of \(2.074\) and an \(R^2\) of \(0.891\) in trust-score prediction. The ABAC layer achieves \(93.27\%\) accuracy, with a FPR of \(3.88\%\), validating its practical feasibility within realistic DER environments.

0.32

A Matched Evaluation of QUBO Action Models and Defensive Remediation in Cyber Attack Graphs

cs.CR ·2026-11-01 ·Kasymkhan Khubiev, Yaroslav Kholodov, Alexey S. Petrenko
Abstract · 20 pages, 3 figures

This study examines how vulnerability-specific actions affect quadratic unconstrained binary optimization (QUBO) models for minimum-cost cyber attack paths and defensive remediation. Edge-flow and variable-length hop formulations are evaluated using simulated annealing, steepest descent, and tabu search against exact classical references. Experiments on six independent synthetic topologies separate the effects of transition eligibility, vulnerability labels, exploit costs, and multiple exploit alternatives. A single labeled action per eligible transition preserves feasible paths, optimal costs, and QUBO size. Under the tested settings, multiple exploit alternatives were associated with approximately unchanged valid-path recovery, lower optimal-path recovery, and substantially larger representations, particularly for the hop formulation. These recovery differences remain descriptive because they are not statistically significant after correction for multiple comparisons. An evaluation on a published 3CX attack graph with synthetic costs demonstrates applicability beyond the generated topologies. The defensive extension selects patches within a budget to raise attack-path costs to specified thresholds or disconnect protected targets. Against exhaustive and mixed-integer references, the tested QUBO heuristics recovered an optimal patch set in 12 of 16 calls across eight scenarios on one topology. The results characterize formulation correctness, solution recovery, and resource requirements using classical local optimization, without establishing quantum computational advantage.

0.29

SANAI: Secure ECC Distributed Keyed Parity Embedded in Weights for Protecting Neural Networks against Bit-Flip Attacks

cs.CR ·2026-11-01 ·Habibur Rahaman, Daniel Lobo, Md Rezoan Ferdous, Swarup Bhunia
Abstract

Quantized neural networks store weights in dense memory, where bit flips caused by voltage scaling, aging, radiation, or deliberate attacks such as RowHammer can severely degrade model behavior. A single corrupted high-order bit can cause misclassifications or nonsensical language-model outputs. Embedded error-correcting codes (ECC) avoid dedicated parity storage by hiding parity within model weights, but existing schemes embed uniformly across layers and bit planes, leave parity unsecured, and tolerate bit-error rates (BERs) only up to about 6%. An adversary aware of sensitive weights and the ECC scheme can also bypass protection by exceeding the correction capability of targeted codewords. We propose SANAI, a sensitivity-guided, bit-plane-selective scheme that protects critical model weights against bit flips with zero additional parity-storage overhead. SANAI secures parity through keyed embedding across a two-dimensional weight space: spread-transform dither modulation binds parity to a secret key, and keyed interleaving disperses concentrated attacks across codewords. We derive a survival law showing that robustness is governed by the correctable fraction t/n rather than by t alone, and a closed-form bound on the protectable fraction of a model. As a streaming checker on an 8x8 systolic array, SANAI verifies weights on the load path with 0.41% area overhead and no additional memory. Across ten vision backbones, three quantization schemes, and ten language models from 82M to 13B parameters, SANAI stays within about one percentage point of clean accuracy at BERs up to 12%, twice the maximum of prior embedded-ECC approaches. It survives fine-tuning bit-exactly, neutralizes progressive and targeted bit-flip attacks, and exactly corrects AttentionBreaker and SilentStriker attacks.

0.28

One String to Rule Them All: Measuring Consent Under the IAB Global Privacy Platform

cs.CR ·2026-11-01 ·Muhammad Abu Bakar Aziz, Christo Wilson
Abstract · 11 pages, 8 figures, 3 tables

Many privacy regulations, including the California Consumer Privacy Act (CCPA), require websites to honor users' opt-out choices. To facilitate this process, industry has introduced technical standards such as the IAB Global Privacy Platform (GPP), which aims to standardize collection and transmission of user consent information across multiple jurisdictions. However, it remains unclear whether these standards consistently and correctly transmit user opt-out choices, or whether opt-out choices are respected by advertisers. In this study, we conduct an in-depth analysis of consent propagation under the GPP framework. We collect data from four web crawls across four experimental settings to analyze how widely the standard is adopted, the extent to which opt-out signals propagate from publishers to third parties, and whether these signals reach and alter third-party advertiser behavior. Our analysis reveals several inconsistencies in the implementation and adoption of the standard that result in user consent signals being lost and user opt-out choices not being respected by the advertisers.

0.28

Black-Box Adversarial Example Attacks on Machine Learning based Election Systems

cs.CR cs.AI ·2026-11-01 ·Subek Acharya, Syed Ayon, Khaleque Md Aashiq Kamal, Alina Barnett, Surya Eada +1 more
Abstract

Machine learning models have shown exceptional accuracy for ballot classification in election systems. Recent research has demonstrated that these ballot classification models are vulnerable to adversarial example attacks. This is a critical security issue because a successful attack could alter the outcome of an election and undermine the democratic process. However, all adversarial example attacks in the voting domain have assumed a white-box threat model, where the classifier parameters are known. In this work, we advance the field of machine learning in election systems by analyzing a realistic black-box adversary. We implement and evaluate eleven state-of-the-art black-box attacks: query based attacks ADBA, Square, SurFree, RayS, and SA-MOO. We also test adaptive transfer based black-box attacks with APGD-$l_\infty$, APGD-$l_2$, APGD-$l_1$, PGD $l_0$, PGD $l_0$+$l_\infty$, and PGD $l_0$+$σ$. Our attacks span multiple different norms ($l_\infty$, $l_2$, $l_1$, and $l_0$) and three different black-box adversarial threat models (decision based, score based and transfer based). In addition to digital analyses, we physically print and scan $20,000$ adversarial examples to evaluate the real-world performance of different black-box attacks. Our finding demonstrate an even larger physical-digital robustness gap exists for black-box adversarial examples than for white-box adversarial examples. Lastly, we conduct the first ever adversarial example transferability study in the voting domain and test adversarial training and barrier zones as two possible black-box defense strategies.

0.27

Reddit Community Signals as a Supplementary Domain-Reputation Feature in Spam Classification: A Conservative Controlled Evaluation

cs.CR ·2026-11-01 ·Fatima Salcedo, Jide Edu
Abstract · 25 pages

Email spam and phishing remain persistent threats, and purely algorithmic classifiers may struggle with unfamiliar senders, evolving message content, and cold-start domains. This study evaluates whether Reddit-derived community signals can serve as a supplementary domain-reputation feature in email spam classification. We use a preprocessed SpamAssassin-derived corpus of 5,582 emails and compare three feature configurations: (1) a baseline using TF-IDF and structural features; (2) a traditional-enhanced configuration adding urgency and action scores; and (3) a Reddit-enhanced configuration adding a binary discussion-presence indicator, a keyword-based Reddit score, and an urgency-Reddit interaction term. We train Logistic Regression, Random Forest, and Support Vector Machine models. In the preliminary single-split evaluation, the Random Forest model with Reddit-enhanced features achieves 97.94% accuracy, a 0.55 percentage-point improvement over the baseline, although the improvement is not statistically significant in the primary test split. Logistic Regression performance declines when additional features are introduced, and the SVM performs poorly and is treated as a negative control. Reddit signals appear most informative for rare domains, but approximately 70% of emails receive no Reddit-derived signal. Because the Reddit data are temporally later than the email corpus, we do not interpret the results as evidence of real-time detection. Instead, the study provides a conservative evaluation of community-derived signals as a supplementary, asynchronous domain-reputation feature and identifies the additional validation required before production deployment, including domain-grouped validation, temporally aligned contemporary data, and adversarial robustness testing.

0.27

Beyond Prompt Injection: The Tool Update Gap in LLM Agent Security

cs.CR cs.AI ·2026-11-01 ·Shiyu Kan, Yue Qi, Yufei Han, Xinjian Luo, Guojun Xiong +1 more
code/data
Abstract · 12 pages, 5 figures

Large language model based agents increasingly rely on external tools to invoke services and interact with web resources. These tools can evolve through software updates, yet agentic frameworks may continue trusting them without re-verifying their safety after implementation changes. This tool update gap creates a supply-chain vulnerability in agentic systems: because updated tool code executes outside the LLM's context window, malicious functionality embedded in tool updates can bypass defenses centered on model-visible interactions, leaving agents exposed to stealthy, persistent threats that existing in-context defenses can hardly address. To thoroughly investigate this threat, we introduce ETUDE (Exploiting Tool Updates for Direct Execution), an attack that embeds malicious functionality into updates of initially benign tools to produce compromised tools, while preserving their legitimate tool functionality. Unlike prompt-based attacks, ETUDE operates entirely outside the LLM's context and requires no adversarial instructions, thereby bypassing defenses restricted to model-visible content. We evaluate ETUDE across three representative attack workloads and seven widely used commercial and open-source agentic frameworks, including Gemini CLI, Codex CLI, and Claude Code. The results show that ETUDE achieves a 97.6% ASR while maintaining a 95.9% legitimate task completion rate. We further examine existing defenses and find that they provide insufficient protection against this threat. These findings expose a fundamental blind spot in current agent security practices: trust established for a tool does not necessarily remain valid after its implementation updates.

0.26

StateGate: A Behavioral-Fidelity Prompt Optimization Method for Long-Chain LLM Agents

cs.CR ·2026-11-01 ·Shicheng Huang, Cai Fu, Jianqiang Lyu, Yifan Zhang, Xianmao Ji
Abstract · 13 pages, 3 figures, 7 tables. Preprint

Currently, long-chain LLM agents commonly suffer from behavioral degradation in complex tasks, where prompts are written in full but steps are not executed in full. To address this issue, existing approaches face three major challenges. First, constraints are strongly bound to their textual carriers; once natural-language constraint sentences are removed, their binding force vanishes. Second, the execution process lacks independent verification, as the evidence for judgment is often supplied by the executing party itself, leading to the failure of self-referential validation. Third, existing remediation schemes are difficult to generalize, and hard-coded mechanisms cannot be transferred across tasks. In this paper, we propose StateGate, a novel prompt-optimization scheme that transforms semantic constraints into symbolic gates enforced by an external deterministic runtime. By decoupling the content of a constraint from its carrier of efficacy, StateGate achieves behavioral fidelity in long-chain tasks without losing the binding force of the prompt. Specifically, we first propose four independent compression strategies (deduplication, generalization, externalization, and text-to-gating) together with their corresponding decision chains, enabling principled prompt symbolization. Second, we design a three-component architecture consisting of a state file, a validator, and an orchestrator, which thoroughly separates judgment authority from execution authority. Third, we formally prove that the compliance of symbolic constraints is independent of context size. Finally, we validate the framework on a vulnerability-discovery task in the security domain; the results show that StateGate achieves 100% execution closure and accurate vulnerability detection.

0.15

Query-efficient scan-and-search watermark localization

cs.CR cs.AI ·2026-11-01 ·Liyuan Wang, Xinsong Ma, Ziye Ma
Abstract

Recently, the detection of AI-generated content has been of particular importance in various fields of our society. For accountability purposes, many large language models (LLMs) have implicitly watermarked their outputs. Localizing such watermarks requires a piece of text to undergo an additional verification process. Existing localization methods typically rely on dense token-level scoring, making verification costly even when an LLM-generated watermark only sparsely appears. To tackle this problem, we design a new adaptive allocation strategy that first maintains broad document coverage and then progressively concentrates the detection budget on regions that exhibit stronger watermark evidence or are more likely to contain watermarked content. We then transform the statistical signals from the selected regions into e-values and employ the proposed token-weighted e-BH procedure to determine the localized watermark regions. The resulting framework is called BELL (Budgeted E-evidence Localization for LLM Watermarks). Theoretically, BELL can control the token-weighted false discovery rate (FDR) without independence assumptions across regions. Finally, extensive experimental results demonstrate that BELL consistently outperforms competing methods across multiple LLMs, datasets, and watermarking schemes.

0.12

CertClaim: Towards Certified Model-Claim Verification for Coding-Agent Services

cs.CR ·2026-11-01 ·Jiacheng Lu, Qixuan Yang, Tao Song, Daxin Jiang, Jing Li +2 more
Abstract

Coding-agent services are priced by the model they claim, making the verification of such model claims crucial for their users. In this paper, we reveal that existing model auditing methods rely on a latent assumption that the deployment is static, i.e., the claimed model is always served in one reference configuration. This assumption does not hold for coding-agent services, whose providers may legitimately change the sampling preset, the reasoning mode or the system prompt without changing the model. As a result, existing methods either falsely accuse honest providers or miss cheap derivatives of the claimed model. To address this problem, we propose a certified model-claim verification method (dubbed CertClaim). Instead of testing the service against one reference deployment, we make the authorized deployments part of the claim and test whether some authorized deployment explains all the evidence of the service, by calibrating multiple evidence channels in every authorized deployment and fusing them within each deployment. We provide theoretical analyses of its false-accusation rate and detection condition. Extensive experiments on eight claims, two agent harnesses and three task suites verify the effectiveness of CertClaim, which is valid in every setting while separating substitutes far better than existing methods, as well as its resistance to potential adaptive attacks.

0.10

Systems & Networking 29

Wireless Networks That See: Integrated Sensing and Communication in Complex Environments

cs.NI ·2026-11-01 ·Xingqin Lin
Abstract · 7 pages, 6 figures

The same radio interactions that complicate wireless communication can reveal what is happening around a network. Integrated sensing and communication (ISAC) turns these interactions into opportunities to observe unmanned aerial vehicles (UAVs), road vehicles, and industrial mobile robots. Yet a strong echo is not necessarily a reliable object observation: stationary background can conceal motion, reflections can create virtual objects, and one viewpoint can leave part of an object's motion unobserved. This tutorial explains how known cellular signals are used to estimate propagation, suppress stationary background, interpret reflected paths, and combine observations from different sites. UAV examples under a Third Generation Partnership Project (3GPP) stochastic channel model and ray tracing in a San Francisco urban scene illustrate the mechanisms through spatial layouts, channel profiles, and before/after range--Doppler maps. Local checks reduce false outputs, with some loss of detection. Cooperation improves position accuracy for UAVs detected by both a single site and cooperating sites. The examples show how signal resources, measurement reporting, and application needs shape sensing performance, providing practical guidance for sensing services in complex environments.

0.54

Interoperability Assessment of Fronthaul and E2 Interfaces for Multi-Vendor Open RAN Systems

cs.NI eess.SP ·2026-11-01 ·Farhad Mehran, Charles Turyagyenda, Dritan Kaleshi
published
Abstract · Proceedings of European Wireless 2025, Sophia Antipolis, France, pp. 91-96. Published by VDE VERLAG

Open Radio Access Networks (RANs) aim to transform the existing wireless network architectures by adopting open interfaces, disaggregating and virtualising RAN components, and enabling dynamic control of the system via data-driven methods. To accelerate the development of interoperable RAN solutions, rigorous assessment of Open Fronthaul and E2 interfaces based on O-RAN Alliance Interoperability Testing (IOT) specifications plays a key role as the integration of these interfaces seems to be very challenging, e.g. due to the need for alignment of Fronthaul IOT profiles and unforeseen challenges for E2 interface integration as Near-Real-Time RAN Intelligent Controller (Near-RT RIC) is relatively a new component introduced within the RAN architecture. In this paper, selected parts of the latest findings and lessons learned from interoperability assessment of Open Fronthaul and E2 interfaces for 13 multi-vendor Open RAN systems deployed in SmartRAN Open Network Interoperability Centre (SONIC) Labs are presented.

0.53

Superscalar Loop Processors

cs.AR cs.DC ·2026-11-01 ·Luca Colagrande, Pascal Etterli, Luca Benini
top venuecode/data
Abstract · Accepted at MICRO 2026

To sustain the computational demands of modern ML and HPC workloads, programmable accelerators integrate large arrays of processing elements (PEs), typically built around single-issue in-order cores to maximize integration density and energy efficiency. As a result, each PE's instruction throughput is upper-bounded at one instruction per cycle (IPC), often suffering from significant underutilization of its functional units (FUs). Conventional superscalar processors can improve FU utilization by increasing IPC, but they typically incur substantial area and energy overheads. This work proposes the Superscalar Loop Processor (SLP) architecture, a lightweight approach to superscalar execution that targets regular, compute-intensive loop kernels. We present the SLP execution model, which exploits the static structure of these workloads to enable efficient multiple-issue execution without the complexity of dynamic out-of-order scheduling. We demonstrate the SLP architecture through an open-source prototype implementation, Schnizo, developed as an enhancement of a tiny RV32I core with floating-point support. Across several ML and HPC microkernels, Schnizo achieves substantially higher IPC and FPU utilization than the scalar baseline while retaining the efficiency of lightweight in-order designs. These results demonstrate that SLPs can provide an efficient architectural solution for programmable accelerators targeting regular, compute-intensive workloads.

0.51

Token-as-a-Utility: A Model-Agnostic Architecture for Metering and Provisioning AI Inference Services

cs.DC cs.NIcs.PF ·2026-11-01 ·Yichiet Aun, Soung-Yue Liew, Yu Liang Lai, Jing Jie Tan, Dong Chen +1 more
theory
Abstract · 13 pages, 7 figures; 41-page supplementary material in ancillary files. Submitted to IEEE Transactions on Services Computing. Data and code:…

Token-based subscriptions are emerging as a commercial model for large language model inference, but extending them into interoperable utility services requires comparable consumption accounting. Raw token counts do not consistently measure inference resource use because demand varies with model architecture, context length, tokenisation, and concurrent serving conditions. This paper presents Token-as-a-Utility, a model-agnostic architecture integrating inference metering, service-level-aware provisioning, and inter-operator settlement. We prove that, under batched serving, no per-request meter can be additive, independent of co-resident traffic, and exactly cost-reflective at once. This motivates separating provisioned capacity, metered consumption, and contracted quality into distinct service products. Consumption is expressed in Reference Inference Units (RIUs) derived from prefill and decode demand, while occupancy-indexed operator ratings recover shared serving costs without altering the common unit. Versioned inference records support auditing and re-rating across operators; software-enforced capacity leases and quality-constrained routing enable provisioning on shared accelerators. The resource model is measured on three NVIDIA RTX 5000 Ada GPUs, and multi-tenant behaviour is evaluated with a trace-driven model calibrated to them. For the reference model, occupancy-indexed rating limits statement energy error to 9% under measured traffic, compared with up to 70.5% under fixed-occupancy rating. In calibrated multi-tenant experiments, tenant mispricing decreases from 7.36 times under flat token counting to 2.39 times with RIUs and 1.06 times after occupancy-based reconciliation. These findings demonstrate how portable inference accounting can be combined with load-dependent cost recovery and contract-based provisioning to support heterogeneous inference services across operators.

0.42

Naviq: An Open-Source System-on-Chip Simulator for Rapid Prototyping, Verification, and Design Space Exploration

cs.AR ·2026-11-01 ·Scott Smith, Lukas Zscherpel, Alyssa Huang, Marissa Lanz, Deming Chen
publishedcode/data
Abstract · 9 pages, 10 figures. This work is accepted to ASPDAC'27

Modern FPGA SoCs are no longer just collections of accelerators on top of a programmable fabric. They combine programmable logic with hardened processors, NoCs, and components that dominate end-to-end performance. Evaluating individual pieces is insufficient; system performance depends on the interplay of SoC components. However, current FPGA SoC simulations provide limited support for this style of evaluation. Vendor simulations capture these interactions but are prohibitively slow for iterative design-space exploration. At the same time, generic software models overly abstract the hardware components that determine real behavior, resulting in less accurate simulations. Fast, easy-to-use simulators that accurately model real SoCs are critical for system debugging and design space exploration. We present Naviq, an open-source, cycle-level software simulator for FPGA SoCs, tuned to AMD Versal-class architectures. Naviq bridges accurate but slow vendor simulation and fast software modeling: it captures key behaviors of modern commercial FPGA SoCs while being fast enough for debugging and design-space exploration. Naviq models hardened SoC infrastructure in C++ and integrates accelerators through Verilator, allowing designers to evaluate custom hardware in a full-system environment. At the center of Naviq is an AXI4-compliant NoC model that captures transaction ordering, packetization, routing, backpressure, and memory interactions. Naviq provides CPU and memory integration, protocol checking, checkpointing, tracing, and performance-analysis tools for debugging and design-space exploration. Experimental results show Naviq closely reproduces the bandwidth and latency of Vivado's RTL NoC model while accelerating the evaluated simulations by 81x-182x. We present five case studies demonstrating Naviq's capabilities. Naviq enables practical full-system evaluation of modern FPGA SoC designs.

0.38

Federated Vector Retrieval across Heterogeneous VectorDBs

cs.DB cs.IR ·2026-11-01 ·Mintaek Lim, Sieun Hyeon, Geon Chio, Jaeyoung Do
top venuebenchmark
Abstract · Accepted for publication at VLDB 2027. The manuscript contains 13 figures

Vector databases are now core infrastructure for AI retrieval systems, yet many real deployments must serve queries over multiple independently constructed vectorDBs built with different embedding models. This creates a systems problem of federated retrieval over heterogeneous vectorDBs: vectors and similarity scores are not directly comparable across embedding spaces, while fully resolving this incompatibility through corpus-wide re-embedding and re-indexing is often operationally prohibitive. Existing methods therefore face a difficult trade-off between retrieval quality, construction cost, and query throughput. We present HVMap, a lightweight compatibility layer for federated retrieval across heterogeneous vectorDBs. Instead of rebuilding databases into a unified space, HVMap learns bidirectional mappings between embedding spaces using a fuzzy contrastive loss, enabling calibrated cross-database scoring without corpus migration or index reconstruction. HVMap is system-agnostic and further scales to multi-vectorDB settings while preserving ranking fidelity through a hub-based architecture. We evaluate HVMap in open-domain, cross-domain, and scale-out settings, and also construct a new benchmark for heterogeneous cross-domain federation. Across these settings, HVMap consistently achieves a better quality--cost--throughput trade-off than baselines, achieving up to 3.59X higher nDCG@10, up to 2.01X higher query throughput, and construction time as low as 0.3% of baseline methods. These results show that HVMap makes heterogeneous vectorDB federation practical at scale.

0.37

QFabric: Guarded Cross-Domain Task Recovery and Its Limits on a Linux-RTOS Board

cs.AR cs.OSeess.SY ·2026-11-01 ·Ayushman Bhattacharya, Anwesha Chakraborty
Abstract · 9 pages, 4 figures, 8 tables. Code and data: https://doi.org/10.5281/zenodo.23692354

Heterogeneous embedded boards pair a Linux application processor with a real-time microcontroller (MCU), which suggests moving a task between the two domains when its timing degrades. We present QFabric, a prototype that treats cross-domain placement as guarded recovery from a violated empirical soft real-time contract: a bounded task is compiled for both domains, monitored in windows with hysteresis, switched only after semantic and capacity admission at a quiescent epoch boundary, and committed or rolled back after probation. On an Arduino UNO Q (Cortex-A53 Linux and a Cortex-M33 running Zephyr), we characterize the stock Bridge RPC interconnect and measure five workloads in both domains under idle, CPU-load, and higher-priority SCHED_FIFO contention, with ordinary and real-time callers. Each MCU invocation costs about 19 ms of dispatch regardless of compute, the MCU runs the heaviest kernel 34x slower, and because every call traverses Linux userspace, Linux contention inflates the MCU path far more than the Linux path (p99 up to 400 ms versus 17.0 ms). Across 30 cells of a four-node chain, Linux meets every contract within 0.8% misses and QFabric correctly never migrates; a controller without hysteresis migrates in 7 cells, raising p95 by up to 177x, and probation rolls back 5 of those moves. On Linux-dispatched heterogeneous boards, migration therefore cannot isolate a task from Linux interference, and recovery guards matter chiefly by preventing harmful moves. This paper supersedes an earlier artifact release whose reported improvement arose from measurement defects that we identify. All source, raw traces, and generators are public.

0.37

Graphs Have Structure: Declarative Manifolds, 3D Z-Order Monoid Roll-ups, and Zero-Simulation Visual Analytics in a Graph DBMS

cs.DB cs.CGcs.HC ·2026-11-01 ·Madhulatha Mandarapu, Sandeep Kunkunuru
code/data
Abstract · 7 pages, 1 figure, 4 tables. Companion to arXiv:2606.24677 (OEH). Code: https://github.com/samyama-ai/structural-manifold-index and https://…

Every real-world graph is embedded in an ambient structural manifold: proteins fold in Euclidean space ($R^3$), telecom networks sit on a planetary surface ($S^2 \times R$), and ontologies and calendars form discrete subsumption posets ($(V, \sqsubseteq)$). Yet property graph databases treat node coordinates as opaque payload scalars, leave spatial proximity edges unindexed, and force visualizers into $O(V^2)$ client-side force-directed simulations that destroy true geometry and truncate hierarchy roll-ups. We elevate ambient manifold structure to a first-class primitive in Samyama Graph. First, we introduce a declarative and auto-detected structural convention contract (samyama.structure/1) persisted in .sgsnap headers that binds node labels to Euclidean 3D, Geo-Spherical, Causal Spacetime, or Poset Hierarchy manifolds. Second, we show that both discrete posets (via Order-Embedded Hierarchy $[tin, tout]$ DFS intervals) and continuous 3D manifolds (via a 63-bit 3D Morton Z-order Structural-Geometric Index, SGI) reduce regional aggregation to 1D rank-interval monoid folds answered directly from index-resident Fenwick trees (SUM, COUNT) and Segment Trees (MIN, MAX). At $100,000$ 3D nodes, SGI builds in $28.14$ ms and answers Morton-interval SUM and MAX roll-ups in $0.048\,μ$s and $0.104\,μ$s---$8,345.1\times$ and $3,824.4\times$ faster than property scans. Third, we introduce 3D voxel-hashed virtual contact edges weighted by Topological--Geometric Stress $σ(u,v) = d_{graph}(u,v) / d_{manifold}(u,v)$ to isolate long-range tertiary folds, plus manifold-guided $A^*$ shortest-path search that cuts explored states by up to $3.80\times$ over Dijkstra. Fourth, we replace client layout simulation with server-computed 3D Cone-Tree/Ribbon coordinates and a non-WebGL 2.5D Isometric projection with painter's-order depth sorting and sub-millisecond Level-of-Detail (LoD) macro-cuts.

0.35

Vehicle and UAV Detection and Tracking with LTE Downlink Reference Signals: A Ray-Traced Evaluation on a Mapped Suburban Site with Licensed Base-Station Records

cs.NI ·2026-11-01 ·Arun K. Gurung, Satha K. Sathananthan
Abstract · 18 pages, 7 figures, 14 tables. Revised version

Passive sensing on the cellular downlink places the receiver at the user equipment and changes nothing in the network. We report simulation results for an LTE band-3 cell-specific reference signal arm on a ray-traced channel. On a synthetic street canyon the arm is compared with an uplink sounding reference signal arm on the same band, target and scorer: missed-detection probabilities of 0.667 and 0.772. On a surveyed site built from OpenStreetMap footprints and the licensed site record, nine receiver positions along the street track the vehicle at coverage between 29.9 % and 61.7 %; a 283 m lap separates target height from trajectory at 1.6 to 30 m; a 137 m lap on the built-up street tracks a vehicle and a 15 m airborne target on every cell, and 8 receivers at 135 m spacing carry one illuminator's coverage over 1140 m of the street. Potential future work includes extending the analysis to multiple illuminators and varying receiver configurations, and confirmation through over-the-air measurements.

0.34

Layer-wise CPU-GPU Scheduling for LLM Inference on Memory-Limited Consumer GPUs

cs.DC ·2026-11-01 ·Naoto Sugiura, Yuta Tokusashi, Hiroki Matsutani
published
Abstract · Euro-Par 2026

As large language model inference environments expand toward consumer devices, it remains challenging to efficiently run models that exceed GPU memory capacity in single-GPU settings. The latency of transferring weights that do not fit in GPU memory accounts for about 98\% of inference time and remains a major obstacle. In this paper, we focus on the fact that during the decode phase, layer computation time on the CPU can be shorter than weight transfer time, presenting an opportunity to hide weight transfers by utilizing this computation time. Therefore, we propose a method to hide transfer latency by executing certain layers on the CPU during the decode phase while asynchronously transferring the next layer's weights to the GPU in the background of that computation. To realize this scheduling, we formulate the placement of CPU/GPU computation and weight transfers as a latency minimization problem and perform optimization using an approximation model with relaxed temporal constraints. Evaluation results confirm that in a consumer desktop environment, our approach accelerates OPT-30B inference by 3.5x compared to HuggingFace Accelerate and 6.3x compared to DeepSpeed Zero-Inference.

0.34

Ascend-SpMM: Accelerating Sparse Matrix-Dense Matrix Multiplication on Ascend NPUs

cs.DC cs.DB ·2026-11-01 ·Jiahe Wang, Yingxia Shao, Shuhan Wang, Qingbo Ge, Hongzheng Li +1 more
code/data
Abstract

Sparse Matrix-Dense Matrix Multiplication (SpMM) is a fundamental kernel in graph neural networks and scientific computing, yet its optimization on Ascend NPUs remains largely unexplored. The physically decoupled Cube/Vector architecture and single-threaded scalar instruction dispatch of modern NPUs introduce unique challenges. We present Ascend-SpMM, a novel and systematic framework tailored to the physically decoupled architecture of Ascend NPUs.We introduce a memory-centric cost model that quantifies hardware affinity and guides a conflict-bounded, densityaware task-mapping strategy. The strategy maps dense regions to the high-throughput Cube path and sparse residuals to the flexible Vector path, with a theoretical bound on cross-processor conflicts. To execute these workloads with high efficiency, we further develop highly customized sub-kernels. These kernels leverage hierarchy-aware pipelining and round-robin ring buffer-based implicit synchronization to effectively mask memory latency and reduce instruction pipeline stalls. Extensive experiments demonstrate that Ascend-SpMM achieves a geometric-mean speedup of 12.70-20.55X over the strongest evaluated NPU baseline for each matrix, accelerates end-to-end GNN training by up to 6.8X, and delivers performance competitive with state-of-the-art GPU implementations. The source code of Ascend-SpMM is publicly available at https://github.com/IDMG-Lab/Ascend-SpMM.

0.32

DIMWIT: Directly Invoking Models Without Intensive Tuning

cs.PF cs.AIcs.CVcs.DC+1 ·2026-11-01 ·Saarth Deshpande, Neeraja J. Yadwadkar, Venkat Arun
hot direction
Abstract

Resource managers typically encode what to do in a task-specific policy---a heuristic, an optimization procedure, or a model trained for the task---that must be developed before use and may need revision as the workload or system changes. We propose bypassing policy development and instead asking pretrained models to make resource-management decisions directly, given the current system state, desired outcome, and available choices. We evaluate this approach on microservice autoscaling, DRAM sizing, and video encoding, using a frontier LLM and two small decision models. On two microservice applications, a small decision model meets latency targets while using within 10% of the replicas allocated by policy-based autoscalers and within 13% of the frontier LLM. In DRAM sizing, a small decision model reduces DRAM usage by up to 27.5% relative to static provisioning while maintaining at least 95% of baseline performance. In video encoding, it completes encodes up to 4.3$\times$ faster than the frontier LLM, with an imperceptible 1.1-point decrease in Video Multi-Method Assessment Fusion (VMAF), a metric of perceptual video quality. Making individual resource-management decisions may be considerably simpler than developing a general strategy for making them.

0.30

Eliminating Servers in State Machine Replication

cs.DB cs.DC ·2026-11-01 ·Dohyun Park, Wenjie Hu, Xiangyao Yu, Mahesh Balakrishnan, Yongjoo Park
Abstract · 14 pages including references, 9 figures, 2 tables

State Machine Replication (SMR) underpins many stateful systems, including distributed databases, coordination services, and metadata stores. Fundamentally, SMR requires state to be materialized from a durable total order of commands. Existing SMR systems rely on servers to materialize, locally store, and serve snapshots of state; while the durable total order either resides on the same servers, or on an external, disaggregated shared log. Unfortunately, servers act as brittle, expensive choke-points in such systems, causing service unavailability when they fail; incurring provisioning latency on cold starts; and red-lining when exposed to bad clients. In this paper, we propose Serverless SMR (S^2MR), where both the materialized state and the durable total order are stored on disaggregated storage; clients concurrently maintain a single copy of materialized state on external storage. Our key technical insight is that the shared log abstraction is useful not just for consistently ordering commands in SMR, but also for enabling distributed materialization, thus becoming the only interface for ac hieving SMR. We demonstrate S^2MR by developing AirDB, a transactional key-value store. In our evaluation, AirD B over S^2MR achieves higher availability than existing SMR-based systems like etcd over Raft: it is oblivious to server failures and bad client activity; and spins up 10x-45x faster from a cold start.

0.27

QueueTrack: Framework for Detecting the Onset of Queueing in Mobile Networks

cs.NI ·2026-11-01 ·Hamid Hassani, Anudeep Karnam, Kishor Joshi, Sonia M. Heemstra, Ignas G. Niemegeers +1 more
published
Abstract

In mobile networks, timely and accurate detection of the onset of queueing at path bottlenecks remains challenging for end-to-end congestion control algorithms (CCAs), where end hosts have limited visibility into queue status. Traditional congestion signals, such as packet loss, round-trip time (RTT), and one-way delay (OWD), have well-known limitations for early queue buildup detection. This paper presents QueueTrack, a practical end-host framework to detect the onset of queueing in mobile networks. QueueTrack computes a queue buildup indicator, denoted QBI, from endpoint-observable packet-level timing statistics, without requiring tight synchronization between end hosts or modifications to the existing network infrastructure. We develop and assess the complete QueueTrack machine learning (ML) pipeline, including data collection, labeling, model training, and testing, on a 5G testbed built on the OpenAirInterface platform, under a range of scenarios with competing TCP traffic. The experimental results show that QueueTrack can reliably detect the onset of queueing at the 5G base station using low-complexity classification models such as logistic regression. For example, QueueTrack achieves an F1 score of approximately 80% in a challenging mixed TCP/UDP scenario. These findings confirm that QueueTrack can serve as a practical framework for developing CCAs that target low delay, as well as for network telemetry and network monitoring in mobile networks.

0.26

Band-Separated Redundant Communication for a Mars-Analog Rover

cs.NI ·2026-11-01 ·Zamiul Rashid, Akid Anis, Tanzim Ahmed, Md. Golam Rabiul Alam, Md. Khalilur Rhaman
unusual
Abstract

Teams in the University Rover Challenge (URC) drive their rovers remotely and often out of sight. The organizers name communication failure as the most common reason a team falls short. A rover therefore needs bandwidth for several camera views and a command path that survives the loss of one radio. We describe and field test the communication stack of the BRACU Mongol-Tori rover, which keeps vision and control on separate radios and bands. Video and RTK corrections travel over a 5.2 GHz bridge with a 2.4 GHz backup, and the Babel routing protocol chooses between them. Every command goes out on two sub-gigahertz radios at once, and the rover uses the backup copy only when the primary falls silent. We tested the backhaul on the bench with a power cut and injected packet loss. We also drove the rover more than a kilometer from the base, and line of sight ended about halfway. On the bench, traffic reached the backup within a few seconds. A metric offset kept it on a primary that lost one packet in ten. In the field, the primary failed at 771 m and the backup carried video and corrections to just over a kilometer. The drive also exposed a limit of Babel, which judges a link only by lost Hello messages. It kept traffic on a primary with a faulty receiver, although the backup reported twice the capacity, and it flapped at the edge of range. We close with lessons on antenna aiming, link metrics and logging.

0.26

PROBE: Learned Proxy-Based Execution for Highly Selective Multimodal Semantic Queries

cs.DB ·2026-11-01 ·Haikai Zhao, Haisheng Fu, Zhengjie Miao, Zhenman Fang
Abstract

Large language models (LLMs) enable semantic queries over unstructured data, but invoking an LLM on every row is prohibitively expensive. Proxy-based execution can reduce this ple with the LLM and training a lightweight proxyto label the remaining rows. Nevertheless, in practical scenarios where semantic queries are highly selective, it remainschallenging to identify representative samples foLM budget and to achieve accurate predictions overthe remaining data using the proxy. We present PROBE, a learned proxy-based execution framework for highly selective multimodal semantic queries. PROBE uses selectivity-adaptive sampling to discover rare polightweight proxy on the acquired labels. It thenrefines proxy predictions through residual correction and homogeneity smoothing over a similarity graph. To address sampling bias, PROBE jointly calibrates refined scores usiarate calibration sample, then selectively routescandidates to the LLM by optimizing the tradeoff between estimated F1 and evaluation cost. On queries from SemBench and LRO-Bench with positive rates as low as 2.17x10^-5 budget yields a geometric-mean F1 of 0.762,compared with 0.797 for exhaustive LLM evaluation, at a geometric-mean speedup of 4.87x and a maximum speedup of 9.3x.

0.24

Apto: Self-Adaptive CPU-GPU Hybrid MoE Inference on Shared PCs

cs.DC ·2026-11-01 ·Zihang Zhong, Jiali Wang, Rong Chen
Abstract · 15 pages, 13 figures

CPU/GPU hybrid inference enables Mixture-of-Experts (MoE) models to run on memory-constrained PCs by keeping all expert weights in host memory and executing activated experts across both processors. Existing hybrid systems, however, assume exclusive resource access and determine an upfront CPU/GPU partition, so CPU contention from colocated applications can severely imbalance the two execution paths. This paper presents Apto, a self-adaptive hybrid inference system that applies dynamic scheduling to expert execution. The CPU and GPU executors claim experts from opposite ends of a dual-ended queue as they progress, allowing the CPU/GPU work split to adapt to their execution progress without predicting contention. Apto further uses affinity-aware scheduling to expose each executor's preferred experts at its queue end and cancellable tail racing to mitigate final-expert stragglers. Compared with state-of-the-art GPU-only and hybrid baselines, Apto achieves average prefill and decode speedups of 2.96$\times$ and 1.52$\times$ under standalone execution, and 5.97$\times$ and 3.45$\times$ under concurrent CPU workloads. It also accelerates real-world coding-agent applications by 2.02$\times$ on average.

0.24

SMART-AIMC: Sensitivity-Aware Mixed ADC Resolution with Fine-Grained Tunability for Analog In-Memory Computing

cs.AR cs.AI ·2026-11-01 ·Yung-Chin Chen, Ourania Stouraiti, Naveen Verma
Abstract

Analog In-Memory Computing (IMC) is a promising paradigm for accelerating GEMM-based AI workloads by executing computations directly within the memory array in the analog domain. While IMC delivers substantial energy and throughput improvements, it can suffer from significant task accuracy degradation due to analog-to-digital converter (ADC) quantization noise, which occurs at the GEMM output and is thus different from weight/activation quantization, which occurs at the GEMM input. Mixed ADC resolution allows IMC modules to choose between high- and low-resolution ADC modes with different energy costs, offering a compelling accuracy-energy trade-off. However, selecting ADC resolutions across IMC units while respecting hardware granularity constraints remains challenging. In this paper, we propose SMART-AIMC, a Sensitivity-aware Mixed ADC Resolution framework with fine-grained Tunability for Analog In-Memory Computing. SMART-AIMC correlates the algorithmic task loss to hardware ADC quantization noise in the form of an ADC Channel Sensitivity (ACS) metric. Using this metric, SMART-AIMC systematically, and in a disciplined manner, allocates ADC resources to critical computations. Furthermore, we introduce a sensitivity-based channel reordering scheme that overcomes the hardware granularity constraints with no runtime overhead and approaches the optimal ADC quantization possible for the model. Evaluated across diverse workloads, SMART-AIMC unlocks a superior accuracy-energy Pareto frontier, providing (1) higher task accuracy at matched energy budgets and (2) fine-grained tuning across efficiency targets, from aggressive energy savings to near-digital inference accuracy. At the latter operating point, SMART-AIMC achieves $6.99\times$ higher modeled system-level energy efficiency for linear projections than the INT8 digital baseline with negligible accuracy degradation.

0.23

SparseForge: Beyond Spike Sparsity toward Hardware-Exploitable Neuron Pruning for Efficient SNN Acceleration

cs.AR ·2026-11-01 ·Kainan Wang, Sanhe Fu, Aili Wang
publishedbenchmark
Abstract · Author accepted manuscript. Published in IEEE Transactions on Circuits and Systems I: Regular Papers

Spiking Neural Networks (SNNs) provide an event-driven paradigm for energy-efficient edge computing. However, operation-level sparsity in SNNs does not necessarily eliminate the hardware costs associated with redundant output neurons, including neuron-state maintenance, partial-sum (psum) accesses, and on-chip storage requirements. This paper presents SparseForge, a hardware-aware sparse SNN acceleration framework that jointly optimizes neuron pruning, sparse dataflow, and accelerator microarchitecture. At the algorithm level, we introduce a gated leaky integrate-and-fire (G-LIF) neuron to learn differentiable neuron masks with binary spike semantics. A dual regularization scheme is further developed to align pruning with both synaptic-operation reduction and psum buffer capacity reduction. At the dataflow level, we design a sparse Column-Wise Product (CWP) dataflow with a variant of incremental spike encoding to improve weight-access regularity and reduce sparse-index overhead under dynamic spike activity. At the architecture level, we develop a mask-aware sparse processing unit (SPU) that supports compressed spike decoding, sparse implicit im2col conversion, neuron-mask-gated psum updates, and decoupled sparse execution. Experimental results across multiple SNN backbones and datasets show that SparseForge retains 1.75%--77.99% of neurons and 0.16%--62.36% of synaptic operations, with a configurable accuracy--efficiency trade-off. The buffer-capacity-aware regularization reduces the required psum buffer capacity by up to 96.18\%. FPGA implementations on a Xilinx ZCU9EG achieve 296.94--619.49 GOPS/W across the evaluated workloads. Across matched network and dataset settings, SparseForge provides 1.24--4.17 higher FPS and 1.50--7.18 higher GOPS/W than representative FPGA-based SNN accelerators, while reducing EDP by 1.70--19.13, with benchmark-dependent accuracy--efficiency trade-offs.

0.23

Computation in Anonymous Directed Dynamic Networks: Exact Worst-Case Termination Time and Network-Wise Intractability

cs.DC ·2026-11-01 ·Giuseppe A. Di Luna, Giovanni Viglietta
theory
Abstract · 37 pages, 3 figures

We determine the exact round complexity of Counting in anonymous directed dynamic networks. Agents have no identifiers, but $\ell\geq1$ of them are designated as leaders. Every agent knows $\ell$ and its own leader status, while the population size $n$ is unknown and has to be determined by the agents. Communication proceeds in synchronous rounds; the directed communication graph may change arbitrarily and unpredictably from one round to the next, but is assumed to remain strongly connected. We solve Counting with a distributed algorithm that makes every agent output $n$ and halt after a number of rounds equal to $\begin{cases} n&\text{if }n=\ell\\ 2n-1&\text{if }n>\ell\text{ and }\ell=1\\ 2n-2&\text{if }n>\ell\text{ and }\ell\geq2. \end{cases}$ The algorithm works even for networks with loops and parallel arcs and even when agents learn their outdegree only after sending. We prove that this bound is worst-case tight over directed multigraphs for every $n$ and $\ell$ with $1\leq\ell\leq n$. Moreover, the local computation is polynomial (provided that each communication multigraph has at most $2^{\operatorname{poly}(n)}$ arcs): in every round, each agent solves at most one linear system with up to $n$ variables. By standard techniques, solving Counting also enables the evaluation of any computable function of the multiset of agents' input values within $2n-1$ communication rounds. The bound above optimizes the worst case over all networks. For an individual network, we show that even deciding whether an agent already has enough local information to determine $n$ is $\mathsf{coNP}$-complete (provided that the numbers of parallel arcs are encoded in binary). Thus, worst-case-optimal Counting admits polynomial local computation, whereas exact network-wise optimization is intractable under standard complexity assumptions.

0.21

A Surrogate Model for Variation-Aware Clock-Mesh Analysis and Design-Space Exploration using Graph Neural Networks

cs.AR cs.LG ·2026-11-01 ·Muhammad Hadir Khan, Wajid Ali, Arya Thottiparambil Manikandan, Matthew Guthaus
Abstract

Clock meshes are widely used in high-performance VLSI systems to minimize skew and tolerate On-Chip Variation (OCV) in nanometer technologies. However, analyzing meshes is difficult because reconvergent paths and multi-source drivers prevent static timing analysis (STA) tools from handling mesh loops, making SPICE the industrial standard. SPICE simulations are accurate but slow, so clock-mesh synthesis relies on approximate models that overlook critical effects such as non-linear buffer driving caused by input slew and input skew. We propose a Graph Neural Network (GNN) based surrogate model of the clock mesh with a new graph representation and a new architecture that predicts power while also improving the delay and slew predictions at the sinks and capturing the skew histogram under OCV. Trained on SPICE data and tested on unseen real designs, our model achieves average errors of $3.40$~ps in delay, $4.14$~ps in slew and $4.46$~ps in skew, compared with $6.28$~ps, $12.35$~ps, and $11.62$~ps for our prior model, and $74.37$~ps delay and $27.51$~ps skew for the approximate method used in current state-of-the-art works. Our model predicts power within $3.4\%$ of SPICE on average, while delivering $18\times$-$1706\times$ speed-ups over the approximate method under OCV modeling, enabling faster and more accurate clock-mesh design-space exploration. A case study on a real design further shows that our surrogate model finds buffering candidates that, at a balanced tradeoff, improves $99^{\text{th}}$-percentile skew by $49.1\%$, reduces power by $25.7\%$, and reduces area by $37.3\%$ at the cost of a $6.8\%$ increase in $99^{\text{th}}$-percentile maximum slew as compared to the state-of-the-art work.

0.21

CacheWeave: Learning to Recompute for Efficient Position-Independent KV Caching

cs.DC cs.LG ·2026-11-01 ·Yueyang Pan, Qianchao Zhu, Yuliang Liu, Teng Ma, Sanidhya Kashyap
Abstract · 14 pages

Position-independent caching reuses chunk KV caches across positions and requests. Independent encoding, however, omits cross-chunk interactions and can degrade accuracy. In our study, neither PIC training nor rule-based recomputation, alone or combined, consistently recovers full-prefill accuracy. Oracle-guided recomputation on PIC-trained models improves accuracy at the same budget, motivating better selection. We present CacheWeave, which adapts the full backbone to PIC, then freezes it and trains a model-native block selector on the adapted model's oracles. The selector uses compact cached chunk features and question hidden states without extracting backbone attention scores. It selects contiguous token blocks aligned with KV blocks to support efficient recomputation and limit copy-on-write amplification. Its runtime, CacheWeave-Serve, supports private and shared KV caches, keeps selection and recomputation on the GPU, and stages sliding-window KV for hybrid attention models. Across three models and four tasks, CacheWeave outperforms CacheBlend and ProphetKV by 7.2 and 3.5 percentage points on average at a 5% recomputation budget, with a mean relative accuracy loss of 1.6% to full prefill. With 32-token blocks, CacheWeave-Serve cuts single-request TTFT by up to 16.1x over full prefill and 1.7x over ProphetKV, and raises SLO goodput under load by up to 7.8x.

0.19

Fairway: Load-Aware KV Cache Reuse for LLM Prefill Scheduling

cs.DC ·2026-11-01 ·Hamidreza Alikhani, Nazmul Takbir, Matin Khajavi, Rajat Bhattacharjya, Sangeetha Abdu Jyothi +1 more
Abstract

LLM-based applications often employ agentic workflows and retrieval-augmented generation (RAG), which build long prompts from shared context. When prefixes are shared across requests, serving engines reuse the key-value (KV) cache to skip parts of the prefill computation. Reuse is typically graded: the uncached work of requests in a queue can differ by up to 100x. In this setting, the order of scheduling determines the time-to-first-token (TTFT) for the queued requests. KV reuse-first scheduling, which prioritizes the requests with the longest KV cache hit, gives the lowest median TTFT but starves expensive requests under load spikes; first-come-first-served (FCFS) scheduling bounds the tail but raises the median TTFT. We present Fairway, a prefill scheduler built on adaptive reuse prioritization: it favors KV cache reuse while queues are short, and shifts toward fair allocation as requests wait longer relative to their own prefill time. Fairway employs a virtual dual lane model: a KV reuse-first Fast lane and a Fair lane atop the serving engine's waiting queue that enables speeding up of cheap reuse requests without starving expensive ones. This model automatically adapts to the workload without any per-workload tuning. Fairway lowers the tail TTFT at no cost to the median in transient overload. Across workloads and load levels, Fairway matches or beats the best baseline, with a p99 TTFT up to 8.6x lower than prior prefix-aware schedulers. During load spikes, it keeps the tail of FCFS at a 6x lower median TTFT.

0.19

RegionFence: Adaptive Concurrency Control for Blockchain Parallel Transaction Execution

cs.DC ·2026-11-01 ·Jiaying Feng, Linpeng Jia, Yi Sun
Abstract · Submitted to the Proceedings of the VLDB Volume 20 (for VLDB 2027)

Smart contracts have turned blockchains into general platforms, and their throughput is increasingly limited by transaction execution. Parallel execution can relieve this bottleneck. However, most existing engines apply a single policy within a block, typically optimistic (OCC) or pessimistic (PCC) concurrency control, and thus re-execute or delay whole transactions. Although some database studies have proposed hybrid and adaptive mechanisms, these mechanisms cannot be applied directly to blockchains because the block fixes the serial order, accesses emerge only at run time, and real workloads vary widely. We propose RegionFence, an adaptive concurrency control mechanism for blockchain parallel transaction execution, based on three key designs. First, we propose regions to refine the granularity of control from whole transactions to the parts that touch contended keys. Second, we design fences to decide at run time how each region handles a conflict. Third, we introduce intra-block and inter-block online learning to anticipate likely conflicts. At the same time, the system still produces the same result as serial execution in consensus order. We implement RegionFence in go-ethereum and evaluate it on real Ethereum mainnet blocks from different periods of the past five years. Experimental results show that RegionFence achieves an overall speedup of 9.2x over single-core execution on 64 cores and outperforms OCC by 2.9-5.4x and PCC by 3.9-9.4x in speedup across these periods.

0.18

MATE: Multi-Agent Traffic Engineering for Topology-Aware Distributed SRv6 Scheduling under Bursty Traffic

cs.NI ·2026-11-01 ·Yuang Chen, Shuyi Liu, Chengdi Lu, Fengqian Guo, Hancheng Lu +1 more
Abstract · 16 pages, 15 figures, 5 tables, and submitted to IEEE journal for potential publication

As network traffic becomes increasingly dynamic and bursty, conventional centralized traffic engineering (TE) suffers from long control-loop delays caused by network-state aggregation, centralized computation, and policy distribution. In this paper, we propose MATE, a multi-agent TE method for proactive distributed scheduling in hybrid networks with partial deployment of Segment Routing over IPv6 (SRv6). MATE uses historical traffic stability to characterize the burst risk of each traffic demand and path sensitivity to quantify the ability of candidate paths to accommodate additional traffic, thereby enabling differentiated proactive scheduling tailored to heterogeneous traffic characteristics. MATE adopts an offline-centralized training and online-distributed execution architecture. During training, a shared global critic evaluates joint actions under a reward combining maximum link utilization (MLU) with variability-weighted path sensitivity, while counterfactual difference rewards provide agent-specific credit for actor learning. During online execution, edge agents generate traffic-splitting policies from local observations and apply them through SRv6, avoiding centralized optimization in each scheduling interval. Experiments on the China Environment for Network Innovations (CENI) testbed and simulations on the Sanet topology show that MATE outperforms a multi-agent reinforcement learning (MARL) ablation baseline without agent-specific credit assignment, COPE, and TEAL in most evaluated scenarios and approaches centralized FIGRET on measured traffic traces. On CENI-Abilene, MATE achieves MLU within 10% of the linear-programming optimum in up to 90.1% of test instances. MATE also achieves lower median normalized MLU than FIGRET in most tested failure scenarios, with mean neural-inference times of 2.05--3.14~ms.

0.15

A GPU-Accelerated and Geometry-Aware Quantum-Classical Framework for Optimized PES Scans with CUDA-Q

cs.ET quant-ph ·2026-11-01 ·Panayiotis Christou, Shuwen Kan, Ying Mao
Abstract

Hybrid quantum-classical workloads, particularly Potential Energy Surface (PES) scans in quantum chemistry using batched VQE and QAOA, involve many structurally similar Hamiltonians differing only in coefficients. Existing warm-start and meta-optimization strategies reuse parameters between instances but still consume fixed quantum-classical resources, ignoring cross-instance structural similarities. We present OptiPESS, a GPU-accelerated runtime system for adaptive resource allocation in hybrid quantum-classical execution. OptiPESS introduces the Locality-Occupancy-Variance (LOV) metric, a lightweight, geometry-aware measure of Hamiltonian similarity, coupled with efficient alignment and Pauli-classification passes. Using LOV, OptiPESS performs distance-aware scheduling of iteration and shot budgets, predicting which instances can be skipped, quick-solved, or fully optimized. Implemented atop CUDA-Q with persistent multi-GPU execution and near-linear scaling, OptiPESS reduces classical and quantum evaluations by up to 78.3\% and wall-time by up to 74.8\% while maintaining baseline accuracy. This demonstrates how lightweight static analysis and adaptive scheduling can transform quantum-classical co-execution from fixed-budget loops into a predictive, resource-aware runtime.

0.13

PRISM: RL-Based Rate Control for Low Latency and High Data Rate in Mobile Edge Networks

cs.NI ·2026-11-01 ·Hamid Hassani, Staji Wilson, Kishor C. Joshi, Sonia M. Heemstra, Ignas G. Niemegeers +1 more
Abstract

Emerging delay-sensitive interactive services, such as extended reality and remote operation, demand advanced congestion control algorithms (CCAs) capable of maintaining high data rates while keeping latency low in mobile networks. Existing CCAs typically rely on congestion signals such as packet loss, delay, or explicit congestion notification (ECN) marks, which can be delayed, imprecise, or require extensive upgrades to the network infrastructure. To address these limitations, physical resource block (PRB) utilization has recently been proposed as a reliable early congestion signal in mobile networks [1]. This paper introduces PRISM, a reinforcement-learning-based CCA that uses peak PRB utilization to regulate the send rate under challenging mobility and fading conditions. NS3 simulation results show that, in a mobility scenario where two PRISM flows coexist with two TCP Cubic flows and one TCP BBR flow, PRISM maintains a per-packet mean one-way delay (OWD) of 6.3ms and a $p_{99}$ OWD of 44.6ms, with negligible (0.02%) packet loss, over a 20min high-contention interval. However, over the same interval, the coexisting TCP flows experience 60x higher mean OWDs of 371-390ms and $p_{99}$ OWDs of 563-578ms, with 0.1% packet loss. PRISM achieves a 31x improvement in goodput-over-latency performance, indicating a favorable tradeoff for delay-sensitive applications.

0.11

Exact Shortest Paths and Maximum Flow in Planar Networks in the CONGEST Model

cs.DC ·2026-11-01 ·Sebastian Senge
theory
Abstract

In the CONGEST model, exact single-source shortest paths (SSSP) with nonnegative weights and exact maximum $st$-flow in directed planar networks have both been solved in $\tilde O(D^2)$ rounds, where $D$ is the hop diameter, while the lower bound is $Ω(D)$. Given the approximate distance labels of a recent preprint (arXiv:2606.31192), which we use as a black box, we solve exact directed SSSP in $\tilde O(D)$ rounds. It rests on a lemma that answers queries from a virtual source with the labels, and on a corrected discretization of the reduction of exact to approximate distances (arXiv:2210.16351). All further bounds of the form $\tilde O(D)$ and $D\cdot n^{o(1)}$ use these labels. Through the reduction of arXiv:2303.00811, SSSP with negative weights, or the detection of a negative cycle, takes $D\cdot n^{o(1)}$ rounds with high probability. We weight the face-disjoint graph of arXiv:2411.11718 so that its distances are dual distances, which turns distance problems in the dual into problems in a planar network of hop diameter $O(D)$. This gives exact maximum $st$-flow in directed planar networks in $\min\{\tilde O(D^2),D\cdot n^{o(1)}\}$ rounds with high probability, and in $\tilde O(D)$ rounds with $s,t$ on a common face. In undirected planar networks, the value of a minimum $st$-cut and a minimum cut take $\tilde O(D)$ rounds. For this we repair the auxiliary graphs of arXiv:2608.09500 at $s$ and $t$, and we show that its recursion, as we read it, can miss the minimum cut, also with exact shortest paths: on a network with $11$ vertices it reports $37$ for a minimum cut of $15$. Two more shortest-path computations per level make it exact. The virtual-source lemma, the weighting of the face-disjoint graph and the corrected discretization are machine-checked in Lean 4 with Mathlib; all other arguments are proved on paper only. Numerical checks are not used in any proof.

0.10

Rethinking Model Aggregation for Federated Learning with Domain Skew: A Knowledge-Preserving Perspective

cs.DC cs.AI ·2026-11-01 ·Haotian Wu, Xu Ji, Qiheng Li, Li Xiong, Zibin Pan
Abstract

Domain skew challenges federated learning by requiring a shared model to accommodate distinct visual distributions. Domain-specific experts can capture these distributions separately, creating an opportunity to integrate their representations in a unified predictor. We propose Federated Knowledge-Preserving Learning (FKPL), a two-stage framework that combines domain-oriented expert training with prototype-based classifier calibration. First, a one-shot subspace-based clustering procedure groups clients for intra-cluster training. The resulting expert encoders are concatenated and frozen, preserving each learned feature mapping during global integration. Second, clients interpolate their concatenated features with class-matched domain prototypes and upload the resulting virtual features. The server trains a global linear classifier on these features, concentrating classifier optimization at the server after a single feature upload. FKPL achieves average accuracies of 73.01\%, 67.00\%, and 64.51\% on Office-Caltech-10, PACS, and a four-domain subset of DomainNet, respectively. On DomainNet, its cross-domain standard deviation is 7.50 percentage points, the lowest among the evaluated methods. With the same concatenated extractor, prototype-based calibration improves average accuracy over federated classifier training by 9.99 and 14.94 percentage points on Office-Caltech-10 and PACS. These results identify classifier calibration as an effective way to combine preserved domain representations into a global predictor.

0.07

Software & Programming 19

What reviewers call a Long Method is a length: A pre-registered test of coupling divergence as a god-unit signal, and what it settled instead

cs.SE ·2026-11-01 ·Daniel Sagenschneider
Abstract · 6 pages, 1 table. Ancillary files: the full registered report and the analysis data. Also deposited at Zenodo, doi:10.5281/zenodo.23749773

A companion study found that a complexity outlier rule shows where complexity concentrates but cannot tell a god unit from a complex algorithm. This paper pre-registered a test of one way to tell them apart: weight a unit's complexity by its coupling divergence, the data it hands on minus the data it receives, so that a leaf is set aside. Coupling was measured by resolving every call in 520 MLCQ repositories, 500 fresh Java repositories and 73 Qualitas Corpus systems. The claim fails: the primary, the pooled MCC of the weighted form minus complexity alone against MLCQ's human Blob, is -0.041 [-0.130, +0.054], no difference resolved and underpowered, and the resolved readings go against it. What the data settle is at method level. What MLCQ's reviewers call a Long Method is an absolute length: a constant of 26 lines or more, chosen on the explore half, reaches MCC 0.777 on the confirm half, +0.080 [+0.016, +0.145] over the best per-repository form; every reviewer with ten or more positives, on their non-leaf reviews, has a best cutoff between 17 and 34 lines; and nothing measured inside a method, its calls, variables, data flow or the cost of cutting it, refines the constant. The constant is a limit of human reading. The next form of the construct is the cost of bringing a long method under it, which is where the construct meets architecture.

0.44

Classification of non-Leibnizian Strings

cs.FL ·2026-11-01 ·Furkan Semih Dündar
publishedunusual
Abstract · 12 pages, 1 figure. Accepted for publication in Complex Systems

Leibnizian ideas in foundations of physics has enjoyed a renewed interest. Cyclic string ontology in Wolfram Model multiway rewriting systems has been found to encode finite dimensional quantum systems where what are called as Leibnizian strings have been employed. It has then become important to classify strings in terms of their ability to correspond to a monadic interpretation à la Leibniz. The definition of Leibnizian strings are rather quite non-linear and for that purpose we classify non-Leibnizian strings for which there is a simpler criterion on maximal neighborhoods of characters. If a string is not non-Leibnizian then it must be Leibnizian. We found out three canonical forms of non-Leibnizian strings and noted that two of them are context-free languages whereas one is not.

0.40

Two-Sided Amortized Resource Analysis: Verifying Resource Lower Bounds via Ill-Typing

cs.PL ·2026-11-01 ·Zimu Chen, Di Wang
theory
Abstract · 56 pages. Reproduction artifact: https://doi.org/10.5281/zenodo.23662529

We present a two-sided type system with potential for verifying resource lower bounds of functional programs. Automatic amortized resource analysis (AARA) verifies upper bounds: types carry potential, and typing certifies sufficient resources. A lower bound, however, is a necessity statement---the program evaluates only if enough resources are available---and prior type-based approaches obtain it by switching the substructural discipline of potential, so that upper and lower bounds live in separate type systems. Matching upper and lower bounds characterize a program's cost exactly. Our system combines AARA-style potential with the two-sided type systems of Ramsay and Walpole, whose central guarantee is that ill-typed programs do not evaluate. Typing formulas may constrain arbitrary terms and appear on both sides of the turnstile; the judgment $t : A \vdash q$ states that if $t$ evaluates to a result of type $A$, then at least $q$ units of resource must be available---that is, $q$ is a lower bound. Because potential makes typing formulas non-replicable, our system follows an affine discipline instead of the replicable contexts of the original framework. We interpret types by semantic typing over value states, pairs of values and remaining resources, and we prove soundness on a safety fragment; corollaries recover the classical two-sided guarantees and the soundness of both kinds of bounds in a single proof system. A prototype type checker in OCaml verifies, on a suite of 19 benchmarks, matching pairs of bounds that characterize costs exactly, as well as lower bounds for programs whose upper bounds are unavailable. In a parallel exploration, we study a two-sided linear type system in which every value is a non-replicable resource; there, closures that capture linear resources call for dual typing formulas, and the exact accounting explains the affine design of the main system.

0.36

Phase-wise MLLM Tuning for Multi-framework WebUI Code Generation

cs.SE cs.AI ·2026-11-01 ·Haoran Ma, Linxiao Li, Chenyue Wang, Haochen Sui, Jiechao Gao
top venue
Abstract · Accepted at NeurIPS 2026. 27 pages

Multimodal large language models (MLLMs) achieve strong performance in translating WebUI screenshots into HTML/CSS. However, when they faced multiple frameworks (React/Vue/Angular), they often suffer from negative transfer that leads to compilation failures. These failures commonly stem from mixing framework specific syntax and violating framework constraints. This paper studies supervised fine-tuning for multi-framework WebUI code generation and proposes a phase-wise MLLM tuning method driven by compatibility and heterogeneity signals across frameworks. We first apply a normalization mapping that preserves each framework's native syntax while suppressing noise from variable naming and literal content. We then estimate cross framework heterogeneity and compatibility offline, and automatically derive training groups and their ordering by maximizing a phased objective. Finally, we perform phase-wise adapter tuning to reduce cross framework interference. Experiments on multiple benchmarks show that our proposed method improves both generation quality and compilation success rate compared with baselines.

0.32

Impossible Code, Barely Changed Score: A Replay Study of Judge-Based Evaluation

cs.SE cs.AI ·2026-11-01 ·Masoud Fetanat
code/data
Abstract

Language models are often used to judge generated code without running it. A useful evaluation should reveal substantial changes in the answers it scores. We examine this requirement by repeating an earlier evaluation on 421 coding problems, comparing Qwen-2.5-7B-Instruct with two versions trained using Group Relative Policy Optimization (GRPO). One receives rewards from execution tests (CE-trained), and the other from a strong language-model judge (SLJ-trained). The base model's syntax-failure rate remains stable, but the trained checkpoints produce invalid Python on 63%-67% of attempts. Despite this change, the strong judge's Pass@32 scores fall by only about 3 percentage points, from 96.67% to 93.95% for SLJ-trained. Within the later run, allowing more attempts makes the base and trained checkpoints look increasingly similar under the judge. With 128 attempts per question, the judge accepts at least one SLJ-trained answer on 96.44% of problems, compared with 28.27% under execution tests. We recommend reporting the share of answers that pass Python's syntax check alongside every judge-based score. This check requires no tests or code execution and flags invalid answers that the judge accepts, although it cannot establish functional correctness. Our findings show how a high judge-based score can hide substantial output damage and motivate checking an evaluation's sensitivity to change. Project page and code: https://curiousorbitnotes.github.io/evorobust-replay-audit/

0.31

Eunoia: Area-Proportional Euler and Venn Diagrams in Rust, Julia, Python, R, and JavaScript

cs.MS stat.CO ·2026-11-01 ·Johan Larsson, Nicole M. Lee
unusual
Abstract · 10 pages

Area-proportional Euler diagrams visualize sets and their intersections through regions whose areas represent their sizes. Because many set specifications have no exact geometric representation, constructing these diagrams requires numerical optimization and diagnostics that quantify the remaining discrepancies. Eunoia is a Rust library for fitting area-proportional Euler diagrams and constructing canonical Venn diagrams. It supports circles, ellipses, squares, and rectangles, computes region areas analytically, and provides exact analytical gradients for several shape families. Its fitting pipeline combines multidimensional scaling initialization with local refinement, global optimization when needed, and seeded restarts. Fitted layouts include residuals and goodness-of-fit statistics that help users assess how faithfully a diagram represents the data. Shared plotting routines extract region polygons, place labels, and pack unit glyphs and member names within regions. The same core implementation powers packages for R, Python, Julia, and JavaScript, a C interface, and a web application that fits diagrams entirely in the browser through WebAssembly. By centralizing fitting, geometry, diagnostics, and annotation in an independent library, Eunoia provides consistent set visualization across programming languages and interactive applications.

0.31

Built on Broken Assumptions (BOBA): Why Coding Agents Cannot Be Left Alone

cs.SE cs.LG ·2026-11-01 ·Alejandro Cuadron, Arpit Gupta
code/data
Abstract · 56 pages, 12 figures, 44 tables. Benchmark, data and code: https://github.com/amazon-agi/BOBA-bench

Coding agents are autonomous developers. Given a request, they change the code and report back when done. As such, they can handle not only independent requests but whole chains of them, each request built on the last. Agents are known to report wrong work as done, yet benchmarks still test them one request at a time, on human-written code, checked once. We build BOBA-Bench, 280 chains of 20 consecutive requests drawn from 140 real open-source projects, and evaluate ten coding agents on both. All ten do far worse on chains. On average, they complete 95% of requests taken one at a time, but when requests are chained together, they complete 51%. Worse code is not the cause. Once checked, agent code is nearly as good a base for later requests as human code. The cause is the mistakes agents miss. Coding agents declare a request done once their own tests pass, even when those tests miss a mistake. The next request then builds on broken code and fails about twice as often. The mistake is rarely found later. Nor can agents catch such mistakes by reviewing code. Asked whether an implementation is correct, they are right 53% of the time, barely better than a coin flip. Coding agents are therefore heavily dependent on access to verification. A single external ``that's wrong'' after a failed request and one repair round lift chain completion from 51% to 59%, and a repaired mistake no longer hurts the requests that follow. In simpler terms, coding agents are only as reliable as the checks they can reach: without verification, every request that follows is built on broken assumptions.

0.31

Order-Invariant CFO Contains Local-Order-Invariant FO

cs.LO math.LO ·2026-11-01 ·Derek R. Aoki
Abstract · 19 pages

Order-invariant cluster first-order logic (CFO) and local-order-invariant first-order logic are two fragments of order-invariant first-order logic, both motivated by the study of invariant logics on classes of bounded degree. Our main result is that order-invariant CFO contains local-order-invariant FO on arbitrary classes of finite structures, settling the conjecture of Ghasemi and Grange that order-invariant CFO contains plain FO. By combining this with a collapse result of Ghasemi and Grange, we obtain the equivalence of FO, local-order-invariant FO, and order-invariant CFO on classes of finite colored graphs of bounded degree. The proof makes use of Gaifman normal form, which allows our translation from local-order-invariant FO to order-invariant CFO to be effective. Combined with the fixed-parameter tractability of model-checking for order-invariant CFO on bounded-degree colored graph classes, this shows that model-checking for local-order-invariant FO on bounded-degree colored graph classes is fixed-parameter tractable. To generalize our collapse and model-checking results beyond colored graphs, we use a method based on first-order interpretations which also yields a collapse of order-invariant FO to FO when restricted to graph classes of bounded shrub-depth. Finally, we give an application to the database-motivated frozen-witness logic of Abiteboul, Simon, and Vianu by showing that local-order-invariant FO contains deterministic frozen-witness FO, and thus that deterministic frozen-witness FO collapses to FO on classes of finite structures of bounded degree.

0.30

CrossSWE: Benchmarking Cross-Repository Joint Development for Coding Agents

cs.SE ·2026-11-01 ·Silin Chen, Yuling Shi, Heng Lian, Zihan Zhang, Xiaodong Gu +2 more
benchmarkcode/datacommunity pick
Abstract · Our benchmark is available at https://huggingface.co/datasets/CrossSWE/CrossSWE-Bench

As coding agents achieve increasingly high success rates on software engineering benchmarks, recent efforts have sought to increase task complexity through longer horizons and more extensive code modifications. However, real-world software development often spans multiple interdependent repositories, introducing additional challenges in dependency reasoning, change propagation, and interface consistency. Rather than further increasing complexity within individual repositories, we introduce CrossSWE, a benchmark for cross-repository joint development comprising 171 real-world instances spanning 165 repositories, 20 technical domains, and 13 programming languages. CrossSWE verifies the necessity of coordinated changes by testing whether partial repository modifications are insufficient to achieve the intended behavior. Our evaluation of five LLMs reveals substantial challenges, with the strongest model, GPT-6.1 Sol, achieving only 58.48% Pass@1. Furthermore, cross-repository inconsistencies emerge as the most prevalent failure mode (44.2%), highlighting the difficulty of maintaining consistency across interdependent repositories.

0.29

Extensible Programmatic Specification of Finite Element Methods

cs.MS cs.GRcs.PL ·2026-11-01 ·Theadora Fields Collin, Oded Stein, Gilbert Bernstein, Saman Amarasinghe
Abstract · 52 pages, 25 figures, including appendices

Finite element methods (FEMs) are a powerful and ubiquitous tool for solving engineering problems. Experimenting with different finite elements can improve the quality and efficiency of solutions. In some cases, the wrong (but nonetheless most common) choice of finite element will produce solutions which converge to the wrong answer regardless of mesh resolution. However, in practice, the choice of finite element is not explored due to the complexity of re-deriving and re-implementing finite element methods. Trying a new finite element is challenging because practitioners must manually deduce formulas to use these elements and implement them within a potentially complex system. We address this problem by introducing ElementForge, a finite element system that is parametric over the literate mathematical specification of a finite element in a domain-specific language (DSL). The ElementForge compiler reasons about tensor spaces, tensors, and tensor bases from first principles to derive implementations of finite elements. The ElementForge compiler is able to automatically derive implementations of finite elements previously only derived by hand. Further, ElementForge minimally couples several key mathematical concepts, mainly tensor fields, mesh topologies, sparse tensors, and assembled finite element operators, to produce a complete finite element system that is parametric over the choice of element. Consequently, the elements derived by the compiler can be applied parametrically to new meshes, PDEs, and boundary conditions. We evaluate our system by implementing several simulations with different finite elements, demonstrating that our system can explore tradeoffs in generality, accuracy, speed, and representational complexity. For example, we are able to implement the Morley, Bell, Argyris, and Hermite like elements with less than 50 lines of code and use them all in a single simulation.

0.29

The Scope Grammar of Natural-Language Rules: Counting and Enumerating Readings modulo Logical Equivalence

cs.FL cs.CLcs.LO ·2026-11-01 ·Sherzod Turaev
Abstract · 13 pages, 1 figure, 1 table

A rule written for people fixes its atomic conditions and connectives but leaves open how the connectives are bracketed, and each bracketing is a different logical reading that an autoformalizer must resolve or expose. We study the scope grammar, a deliberately ambiguous context-free grammar whose derivation trees are the readings of a connective skeleton and whose quotient by logical equivalence is what matters downstream. A skeleton with $k$ binary connectives has $C_k$ derivation trees, the $k$-th Catalan number, but far fewer logically distinct readings. For the positive, repetition-free fragment we make the gap exact: through the uniqueness of the read-once normal form we characterize the realizable normal forms by lowest common ancestors and count them with a two-label recurrence over the operator word, verified against exhaustive enumeration up to eleven atoms. The count is invariant under reversal and $\wedge/\vee$ duality; a skeleton is ambiguous if and only if it uses both $\wedge$ and $\vee$, a one-pass test; and the number of classes is at least $2^{\lfloor n/3\rfloor}$. Up to fifteen atoms the alternating pattern is the worst case; its count has a cubic generating function, is the 3-Schröder sequence (OEIS A137842), and grows with base $\varphi^{5/2}\approx3.330$, below the Catalan base 4. We reduce maximality to an inequality on increments, verified to sixteen atoms, and explain why local exchange arguments cannot prove it. We also treat negation, exceptions, and background axioms, where normal-form uniqueness still decides equivalence but counting is open.

0.28

Homomorphisms on a product of free convex semilattices

cs.LO ·2026-11-01 ·Ana Sokolova, Harald Woracek
unusual
Abstract

Convex semilattices form an equational class that combines convex algebras and semilattices by a distributive law. For a set $I$ let $\mathcal FI$ be the free convex semilattice with basis $I$. A product $\mathcal FI\times\mathcal FJ$ is generated by $I\times J$, but is not free. Hence, not every map from $I\times J$ into an arbitrary convex semilattice $X$ extends uniquely to a homomorphism from $\mathcal FI\times\mathcal FJ$ to $X$. We investigate a condition on a map $\varphi$ of $I\times J$ into some convex semilattice $X$ that is always necessary and often sufficient for $\varphi$ to have such an extension to a homomorphism $Φ$ of $\mathcal FI\times\mathcal FJ$ into $X$.

0.27

Semantic Interoperability in Federated Digital Twin Systems: A Systematic Review Through the Levels of Conceptual Interoperability Model

cs.SE cs.AIeess.SY ·2026-11-01 ·Manonmani Sekar, Nasim Nezamoddini
survey
Abstract

When two companies connect their digital twins, software models of their plants kept up to date with live data, each twin must translate the other's data into its own terms, and that translation rule, called a mapping, can stop being correct when one company changes the meaning of its data without telling the other. This PRISMA 2020 systematic review screened 5,833 records and read 119 full texts to determine whether any published system detects such a broken mapping and repairs it while the twins operate. In all 32 included studies the mapping is fixed before operation, and the two ingredients of run-time repair appear only separately: cross-company matching without any check that the match still holds, and detection and repair of changed meaning within a single company. The review turns this gap into six requirements, each illustrated on one running example, that a system claiming run-time semantic interoperability between independently owned twins should meet.

0.24

Trustworthy Execution of Industrial Agents

cs.SE cs.MA ·2026-11-01 ·Zhongjie Wang, Mingyi Liu, Xiang He, Xiao Wang, Shuang Yu +1 more
Abstract · 80 pages, 16 figures, 37 tables

Industrial agents increasingly modify persistent engineering objects rather than merely generate answers. Authentication and successful calls cannot establish whether intended changes occurred, evidence supports a conclusion, or historical results remain applicable. We present a framework for trustworthy industrial-agent execution within a single enterprise governance domain. Eight control domains assign responsibilities across authority, planning, composition, tool execution, actual effects, evidence, baseline and use governance, and recovery and evolution. A four-plane architecture and three control chains connect these responsibilities through typed objects and version-bound interfaces. The specification separates tool returns from actual engineering effects, evidence coverage from engineering judgment, and historical lifecycle state from current applicability. Multi-agent handoffs are an architectural extension examined through deterministic interface tests. Distinct prototypes examine selected mechanisms using conformance tests, ablations, uncertainty stress, CAD/CAE execution, and local persistence. An OpenFOAM fixture produces COMPLETE coverage with FAIL judgment; an expired-job case requires readmission despite completed computation. CAD and recovery counterexamples expose omitted mandatory checks, discarded negative facts, and inconsistent producer-consumer mappings. Targeted repairs support local regression conclusions. The reported 0/29 hazardous commits measure closed-set specification conformance, not a safety probability. We distinguish the complete framework, implemented paths, and empirical evidence, identifying check-inventory completeness and cross-module fact preservation as prerequisites for extending these results.

0.24

Computons for Control-Driven Boolean Circuits

cs.LO ·2026-11-01 ·Damian Arellanes
Abstract · Extended version of the paper "Compositional Control-Driven Boolean Circuits". DOI: 10.1007/978-3-032-22469-9_2

Boolean circuits abstract away from hardware aspects to capture the logical structure and computational behaviour of digital components. Despite their long history of study, their compositional foundations have received comparatively little attention. Building on the theoretical foundations of computons introduced in previous work, we develop a concrete and operationally defined model of high-level computation for control-driven Boolean circuits. We describe colimit-based operators for composing such circuits into sequential, parallel, branching and iterative structures, supported by concrete operational semantics grounded in abstract computing devices. The framework refines the semantics of composition, establishes the algebraic properties of its operators and introduces atomic circuits to support more expressive computation units. We demonstrate the applicability of the proposed framework through examples from propositional logic, biomedicine and electronics, and establish an operational equivalence with respect to classical Boolean circuits through a systematic transformation that accommodates gates of arbitrary arity. By this equivalence, we show that control-driven Boolean circuits are at least as powerful as their classical counterpart. That is, they are capable of non-uniformly computing any Boolean function over inputs of arbitrary length. Implementations in Idris and Java validate the practical realisability of the framework.

0.21

VeriBase: Automated Construction of Theorem Libraries for Software Verification

cs.SE ·2026-11-01 ·Zihao Zhang, Zenan Li, Taolue Chen, Zhe Zhou, Tao Xie +1 more
benchmarktheory
Abstract

Software verification is typically more challenging than coding. Software developers routinely build on mature libraries and APIs, but turning successful verification solutions into reusable libraries still requires substantial expert effort. In this paper, we propose VeriBase, a framework for automatically constructing software-verification libraries by jointly generalizing code and its associated theorems from successful verification solutions. The key challenge is to generalize implementations together with the theorem statements and proofs that depend on these implementations, and reconcile overlapping components across verification tasks. To this end, VeriBase treats code and its associated theorems as declaration families, and designs a three-phase pipeline, which proves seed tasks, generalizes code and associated theorems, and harvests the resulting families into shared interfaces. VeriBase updates dependent declarations as interface change and uses Lean to check the resulting code and proofs. We evaluate VeriBase across four benchmarks with three backend LLMs. The results show that the general-purpose library built from independent 4,191 programs increases the average task success rate by up to 17.0% and meanwhile reduces the proof completion time by up to 21.4%. When VeriBase constructs a benchmark-specific library, further improvement can be observed.

0.20

Modern Software Engineering with Verification Agents: A Spec-First Architecture for Autonomous Synthesis, Verification IRs, and Organizational Scale

cs.SE cs.LO ·2026-11-01 ·Jason Hickey
Abstract

When autonomous coding agents synthesize implementation code, formal specifications, and mechanical proofs simultaneously, the software engineering hierarchy inverts: the formal specification becomes the primary human-reviewed artifact, while source code and proof scripts become disposable build outputs. This paper presents an end-to-end architecture for agent-authored, formally verified software engineering. First, we establish a logic-agnostic foundation combining a spec-first stack inversion, a zero-code-review workflow gated by automated continuous integration (CI) checks, and a minimal Verification Intermediate Representation (VIR) lowered via a thin, verified translator. Second, we pair an extensional Computational Type Theory (CTT) specification layer with Higher-Order Abstract Syntax (HOAS) and linear resource typing in the VIR--eliminating rewrite friction for synthesis agents while defining an experiment, its protocol frozen before the first reader, to measure whether removing boundary coercions improves human specification readability. Third, we scale collaboration across multi-team organizations using Personal Spec-Copilots and a Shared Synthesis Fleet that detects cross-team Semantic Merge Conflicts and fuses modular contracts into zero-copy binaries. We illustrate this architecture on a zero-copy authenticated message router and distinguish our existing empirical baselines (salt, SaltBench, x86lean, and MetaPRL) from newly proposed components.

0.18

RECEIPT: Measuring and Repairing the Return Assumption of LLM Agents in Web Service Operations

cs.SE cs.AI ·2026-11-01 ·Boyang Yang, Ziyao Yang, Yu Zhang, Duo Chai, Xiaoming Chen +1 more
benchmark
Abstract

LLM agents operate Web infrastructure from the command line and rely on a return assumption, namely that each tool call returns within a bounded time and with its requested action finished. Many commands, however, ask a question before they act, and the agent's harness either closes the program's input or keeps it open as a terminal at which nobody types. With closed input, the program can return status 0 without the requested action. With open input, it waits until a time limit, and the agent pays in time and tokens. To the best of our knowledge, this paper is the first to measure and repair the return assumption. We collect 91,469 production tool calls and 89,047 public agent trajectories, test how 8 widely used harnesses launch commands, and build Litmus, a catalog of 158 checked command-line invocations, and Tripwire, a benchmark of 25 task pairs that differ only in whether a question guards the requested action. We then propose RECEIPT, which keeps the input open, reads what each process waits for from system calls, and returns this state to an unmodified agent through a model-checked protocol. On Litmus, closed input causes 11.5% of tested invocations to return status 0 without performing the requested action, and 7 of 8 harnesses close input when no terminal is attached. An unanswered question at a plain terminal multiplies a task's token use by 2.9, and the input connection moves an LLM from 7th to 3rd place among 8. Across 5 LLMs, RECEIPT reduces false claims of success by 73.5% and completes all 25 guarded tasks on a second harness. An agent's reliability therefore depends on what its harness lets it observe as well as on its LLM. Agent builders should report what each tool call did, not only how it ended, and benchmark maintainers should fix and report the input connection before they rank LLMs.

0.18

Launder-Bench: Can Repository-Level Agentic Code Laundering Be Detected?

cs.SE cs.AI ·2026-11-01 ·Yuntong Hu, Hongji Pu, Zimo Li, Ruomeng Li, Xiaotong Liu +3 more
benchmark
Abstract · 46 pages, including references and appendices

Coding agents can now rewrite repository-scale codebases while preserving tested behavior, creating a new challenge for plagiarism detection, license auditing, and software provenance. We ask whether such rewrites can still be traced to their source. Existing clone benchmarks emphasize fragments and bounded edits, while repository-transformation benchmarks evaluate functional correctness rather than provenance. Moreover, ranking the most similar source does not establish derivation when independent implementations can be similarly close. We define repository-level agentic code laundering and introduce Launder-Bench, to our knowledge the first repository-level benchmark for provenance under agentic rewriting, constructed from 282 real-world repositories. A source-hidden reconstruction protocol decomposes source behavior, reconstructs modules and repository structure, and retains rewrites only after execution-based verification. We evaluate the same provenance problem in two complementary settings: Launder-Bench-Pair contains 1,128 pairwise instances asking whether two repositories share provenance, while Launder-Bench-Attr contains 229 source-attribution instances asking which candidate is the source, with abstention when evidence is insufficient. Measured same-domain independent repositories establish how much similarity can arise without shared provenance and provide detector-specific reference bands. Existing methods remain far from saturating Launder-Bench: the strongest static method reaches only 0.66 AUROC on pairwise detection, while repository encoders rank the true source first on up to 22% of attribution queries but earn calibrated attribution credit on only 1-2%. These results reveal a systematic gap between relative code similarity and source-specific provenance evidence, providing a reproducible target for provenance detection under agentic rewriting.

0.15

Theory & Algorithms 43

Every $k$-Quasi-Planar Graph Has Linearly Many Edges

cs.CG math.CO ·2026-11-01 ·Ankan Sadhu
theory
Abstract · 6 pages, 1 figure

For every fixed integer $k\ge2$, we prove that every simple graph on $n$ vertices drawn in the plane with no $k$ pairwise crossing edges has at most $C_kn$ edges, where $C_k$ depends only on $k$. We allow any finite number of proper crossings between two edges, including edges with a common endpoint. For $n\ge3$, removing a small disk around each vertex turns the drawing into a family of arcs on a sphere with $n$ boundary circles. Each arc joins two different boundary circles, and different edges have different unordered pairs of endpoint circles. The arcs are therefore essential and pairwise nonhomotopic, with endpoints allowed to move along their boundary circles during a homotopy. Any family of pairwise intersecting lifts to the universal cover projects to distinct graph edges that cross pairwise. Its size is consequently at most $k-1$. Beeker and Lazarovich's theorem on surface patterns then bounds the number of arcs. Their estimate is linear in the size of a triangulation, and an explicit triangulation of the $n$-holed sphere has linearly many edges and triangles.

0.54

Improved NP-Hardness of Approximation for Max-3-Cut and for Max-k-Cut with Small k

cs.CC ·2026-11-01 ·Ziang Ni
theorycode/data
Abstract · 20 pages, 5 figures. Code, gadget families and certificates: https://github.com/ziangni-sys/max-3-cut

We prove that Max-3-Cut is NP-hard to approximate within a factor of $27/29+\varepsilon \approx 0.931$. This improves the factor $16/17 \approx 0.941$ of Austrin, O'Donnell, Tan and Wright, the best NP-hardness result for Max-3-Cut since 2012. (Their result, unlike ours, also holds for 3-colourable graphs.) We also improve the NP-hardness of Max-$k$-Cut for every $k$ from 4 to 20; for instance, we obtain $45/47 \approx 0.957$ for Max-4-Cut and $33/34 \approx 0.971$ for Max-5-Cut, where the previous factors were 0.987 and 120/121. Our reductions are gadget reductions from Håstad's linear equations over $\mathbb{Z}_3$, with gadgets that have at most six auxiliary vertices. The main new idea is a negation trick: since the negation of an assignment is again an assignment, equations with right-hand sides $t$ and $-t$ can be handled together, and this makes the resulting predicates symmetric under all permutations of the colours. We complement it with a general soundness criterion for such reductions: in effect, the best strategy of the adversary is to use a single colour pattern everywhere. This criterion is the strongest analysis that uses Håstad's theorem as a black box, and, using LP duality, we show that under it no gadget reduction from these equations can beat $27/29$ for Max-3-Cut. Finally, we observe that a classical reduction from Max-3-Cut to Max-$k$-Cut, applied more carefully, already gives $82/85 \approx 0.965$ for Max-4-Cut on 4-colourable graphs, improving the previously stated $85/86$.

0.47

Paying for Fractional Error: Tight Constant-Round Approximation of Covering Problems with Short Broadcasts

cs.DS cs.DC ·2026-11-01 ·Saeed Amiri, Sebastian Siebertz, Wojciech Wawrzyniak
theory
Abstract

We present a general framework for deterministic constant-round approximation of covering problems in sparse networks in the broadcast CONGEST model, where every vertex broadcasts the same $O(\log n)$-bit message to all its neighbors. The framework combines two independent tools. The first is a solver that computes a $(1+δ)$-approximate solution of any local covering linear program with bounded positive integer costs on every class that excludes a fixed minor. The second is a rounding principle: a rounding rule that saves a fixed fraction of its own output, for every feasible input, pays for a solver error fixed in advance, so that an approximate fractional solution yields an exact coefficient. For ordinary domination, we achieve a $2a+1$ approximation against the fractional optimum $γ_f(G)$ for every fixed rational $a\ge1$, whenever every minor has a ratio of edges to vertices of at most $a$. On planar graphs this is seven, matching the lower bound of Hilke, Lenzen, and Suomela. The coefficients are also optimal at bounded treewidth and pathwidth, for $K_t$-minor-free graphs with $3\le t\le9$, on triangle-free planar graphs (five), and on planar graphs of large girth, where it approaches the optimal three. On every fixed surface the exact bound is $7γ(G)$, and we show that a bound of $7γ_f(G)$ is impossible on some fixed surfaces even sequentially. For total domination, on graphs without isolated vertices and under the same minor-density assumption, the exact coefficient is $2a$ against the fractional optimum, including the optimal six on planar graphs. Red-blue domination, and hence set cover and hitting set on planar incidence graphs, has the optimal limiting coefficient four. Our method also gives randomized constant-round and deterministic $O(\log^* n)$-round $(1+\varepsilon)$-approximation schemes, with the randomized cost guarantee holding in expectation.

0.45

Deterministic Polynomial-Time Primality Testing with a Single Binomial Congruence

cs.CC math.NT ·2026-11-01 ·Joseph M. Shunia
theory
Abstract · 12 pages. An earlier version appeared as Revision 2 of ECCC TR23-200: https://eccc.weizmann.ac.il/report/2023/200/#revision2

We give a deterministic polynomial-time primality test whose main verification is the congruence $(1+X)^n=1+X^n$ in $C_n[X]/(X^r-b(Y))$, where $C_n=(\mathbb{Z}/n\mathbb{Z})[Y]/(Φ_r(Y))$. Two identities in the coefficient ring normalize $b$ and ensure compatibility with the main congruence. A finite-field root count and a lattice bound show that an accepted input is a prime power; elementary preprocessing excludes nontrivial perfect powers. We construct $b$ deterministically using Jacobi sums and the uniform Dedekind-zeta zero-free theorem of Cureton and Thorner, which extends OpenAI's Hecke zero-free theorem. The total bit complexity is $L^{32+o(1)}$, where $L=\lceil\log_2 n\rceil$, and verification after constructing $b$ costs $L^{6+o(1)}$. More generally, for a fixed real algebraic $1/2\leθ<1$, a common zero-free boundary $\Re s>θ$ for the relevant Dedekind zeta functions gives $L^{\max\{6,(1-θ)^{-1}\}+o(1)}$. This bound reaches $L^{6+o(1)}$ at $θ=5/6$, and hence under GRH. We state a fixed-base modification of Agrawal's conjecture for odd integers passing the base-$2$ Fermat test, using the modulus $X^r-2$ with $r$ the least positive integer not dividing $n(n-1)$.

0.44

Approximation Algorithms for Maximum-Weight Degree-Bounded Forests

cs.DS math.CO ·2026-11-01 ·Theophile Thiery, Vera Traub, Nicola Widmer
publishedtheory
Abstract · to appear in SOSA 2027

In the maximum-weight degree-bounded forest problem, we are given a graph with nonnegative edge weights and a degree upper bound on each vertex. The objective is to find a maximum-weight forest satisfying the degree bounds. We show that this problem admits a simple $2$-approximation via the greedy algorithm. This already improves over the best-known guarantees obtained via a reduction to weighted $3$-matroid parity, and sets a baseline to improve upon. Our main result is an $\frac{8}{5}$-approximation algorithm for the maximum-weight degree-bounded forest problem. For the special case of uniform degree bounds, our algorithm achieves an approximation ratio of $\frac{3}{2}$. It computes two sets of edges, each of which we then restrict to a feasible solution in an optimal way, and finally returns the better of the two resulting solutions. To obtain the initial two edge sets, we first relax the degree bounds and leverage an algorithm by Lau and Singh [JACM'15] for degree-bounded spanning trees, and second we relax the forest constraints and employ an algorithm for maximum-weight $b$-matchings.

0.39

Second-Order Asymptotics for the Scalar Gaussian Broadcast Channel

cs.IT ·2026-11-01 ·Vincent Y. F. Tan
Abstract · 34 pages, 9 figures

This paper characterizes the second-order coding region of the scalar Gaussian broadcast channel along the curved boundary of its capacity region. The characterization holds under both average and maximal error criteria and allows correlation between the receiver noises. At each interior point of this boundary, the second-order region is a halfspace, as in the work of Scarlett and the author on Gaussian multiple-access channels with degraded message sets. Achievability follows from spherical superposition coding with a blocklength-dependent power allocation. The main innovation lies in the converse. Using Ornstein--Uhlenbeck entropy dissipation and Gaussian functional inequalities, we construct an auxiliary observation through an additive Gaussian noise channel and establish quantitative relative entropy bounds for the channel output distributions conditioned on this observation. The strong functional representation lemma of Li and El Gamal then allows us to select a realization of its independent random seed to obtain a deterministic partition of the retained receiver-1 message set. The component label plays a role analogous to the auxiliary random variable in the capacity characterization, while providing the distributional control needed at the second-order scale. These properties guide the construction of auxiliary output measures whose likelihood ratios admit the cancellations and normal approximation needed to evaluate an information spectrum converse.

0.39

Federated Estimation of Loading Spaces in Tensor Factor Models under Differential Privacy

cs.IT math.STstat.ML ·2026-11-01 ·Zetai Cen, Fan Wang, Haeran Cho
Abstract · 70 pages

Tensor factor models provide low-dimensional representations of tensor-valued observations while preserving their multiway structure. When these observations are held by different clients, privacy constraints may prevent them from being pooled for centralized analysis. We study estimation of common loading spaces in Tucker tensor factor models under federated differential privacy, allowing heterogeneous sample sizes, privacy budgets, and factor strengths. The proposed method combines private stability testing, Huber aggregation, and Gaussian perturbation at each client with weighted aggregation at the server. We establish non-asymptotic upper bounds governed by a scaled harmonic mean of the squared local error rates, and prove minimax optimality up to logarithmic factors under strong factors, fixed factor ranks, and fixed noise variance. We further develop a private calibration procedure that selects the local tuning parameters and aggregation weights without knowledge of the noise level or client-specific factor strengths. The resulting adaptive estimator preserves the privacy guarantee and attains the error rate obtained with oracle tuning. We illustrate the finite-sample performance of the proposed estimators through simulation experiments and an application to a Meditation Traits EEG dataset.

0.38

Optimal Online Learning in Bilateral Trade Through Interest

cs.GT ·2026-11-01 ·Mengfan Ma, Haoyu Song, Yiran Tong
Abstract · 52 pages, 5 figures

We consider online learning in bilateral trade on platforms such as Uber, Airbnb, and eBay. In each round, a platform posts a payment to a seller and a price to a buyer, both with private valuations, and observes only whether trade occurs. Its objective is to maximize cumulative gains-from-trade, the sum of the differences between buyers' and sellers' valuations across completed trades. We measure the platform's performance by its regret against the best fixed common price. Financial self-sufficiency motivates global budget balance (Bernasconi et al., 2024), which allows profits from some trades to subsidize others, provided that the final balance is nonnegative. Such financial planning should also account for the time value of money: profits earned earlier can accumulate interest, providing additional funds for future subsidies. Motivated by this observation, we introduce global budget balance with interest, under which a platform starting with no funds can use both trading profits and accrued interest to finance subsidies. For a horizon of $T$ rounds and a fixed exponent $θ\ge0$, we consider the per-round interest rate $r_T = T^{-θ}$. We study two classic valuation models, adversarial and correlated i.i.d., and obtain tight bounds on regret against the best fixed common price in both. As the interest rate grows, optimal regret in each model drops from $\tildeΘ(T^{3/4})$ to $\tildeΘ(T^{5/7})$. The intermediate rates differ: $\tildeΘ(T^{3θ/4})$ for adversarial valuations when $20/21 < θ< 1$, and $\tildeΘ(T^{(2θ+3)/7})$ for correlated i.i.d. valuations when $1 < θ< 9/8$. Finally, as an extension, we study $α$-regret, with approximation factor $α\ge1$, for adversarial valuations against the best feasible distribution over price pairs, and show that the optimal competitive ratio rises from 2 to 3 as interest strengthens.

0.36

Improved lower bounds for the kissing number in dimensions 49 to 61

cs.IT math.MGmath.CO ·2026-11-01 ·Dmitry Kamenetsky
theorycode/data
Abstract · 9 pages, 4 tables, 1 figure. Code, data and exact certificates: https://github.com/dimkadimon/KissingNumbers

We give explicit kissing configurations in R^d for every d = 49, ..., 61. The lower bound is improved over the best previously published value in all thirteen dimensions, by margins from 232 to 159,220 points and by 506,461 points in total. The configurations are obtained from the antipodal block lift of Yang et al. applied to the 52,416,000-point minimal shell of the extremal even unimodular lattice P_48p, using a class of 7,193 minimal lines and, in each dimension, a kissing configuration of R^k (k <= 15) with the largest available number of antipodal pairs. Two ingredients are new. The published class of 7,077 lines is grown to 7,193 by local search in three stages, as a sequence of exchanges between candidate lines and class lines; and the point count, which equals 52,416,000 + 2S + 2t with t the number of antipodal pairs and S the size of the union of t isometric images of the class, is maximised as a maximum-coverage problem by greedy search with iterated local search over pools of 4,000 automorphic images, reaching 92-100% of the trivial bound 7,193t. All results are verified in exact integer arithmetic, and the certificates are distributed with the code. Code and data: https://github.com/dimkadimon/KissingNumbers

0.36

Worst-User Outage Probability Minimization in Active RIS-Aided Systems

cs.IT eess.SP ·2026-11-01 ·Donghwan Shin, Hyeongtaek Lee, Junil Choi
published
Abstract · 5 pages, 2 figures, accepted to IEEE Wireless Communications Letters

In this paper, a worst-user outage probability minimization problem in active reconfigurable intelligent surface (RIS)-aided systems is addressed. Since the resulting problem involves a discontinuous objective function over an unknown channel distribution, a sequence of reformulations is applied by integrating a probability-simplex user weighting, a softplus surrogate for differentiability, and entropic tail-risk aggregation for rare but severe outage events. Based on the reformulated problem, a data-driven stochastic optimization technique that relies on historical channel samples is adopted to jointly optimize the beamformer and reflection coefficients with projected adaptive moment estimation. Simulation results show advantages of the proposed technique for reliability-oriented active RIS transmission design.

0.36

How Much Does Random Rotation Help GPTQ? Sharp Worst-Case Rate Gaps in the High-Rate Regime

cs.IT cs.LG ·2026-11-01 ·Runguo Li
Abstract · 28 pages, 3 figures

GPTQ is Babai's nearest-plane algorithm on the lattice of the Cholesky factor of the layer's input second moment $Σ$. In the high-rate, entropy-coded regime it needs $\tfrac12\log_2(\AM/\GM)$ of its Cholesky pivots more bits than WaterSIC, which is within $0.255$ bit of the rate-distortion limit for every $Σ$; random rotations close most of this gap in practice, and Ordentlich and Polyanskiy asked for the worst-case gap between rotated GPTQ and WaterSIC. We show that after a Haar rotation the $k$-th pivot converges almost surely, uniformly in $k\geδn$ and over spectra $ν$ of bounded condition number, to $1/S_ν(-k/n)$, a classical formula from the theory of Lyapunov exponents, and use it to answer the question. Over spectra with condition number $κ$ the worst rotated gap is $\tfrac12\log_2κ-\log_2\lnκ+\tfrac12-\tfrac1{\ln2}+o(1)$ bits, attained only by two-point spectra, just $\tfrac12\log_2\lnκ+O(1)$ bits below the worst case without rotation, so rotation cannot replace waterfilling in general. It nearly can on real layers: rotation cuts the cost of an $η$-fraction of outlier eigenvalues of size $μ$ from about $ημ$ to $η^2μ/2$ to leading order, but cannot remove spectral spread on a logarithmic scale; for non-Gaussian weights it also adds a weight-side rate cost, which we characterize exactly under a standard high-rate assumption. On $96$ Hessians of Qwen3 and Llama-3.1 the rotated gap has median $0.04$ bit and maximum $0.36$ bit, against $2$--$5$ bits in the worst case for the same condition numbers, and the limit matches it to within $0.009$ bit.

0.33

Near-Quadratic Error-Free Graph Matching above Otter's Threshold via Approximate Local Tree Correlation Testing

cs.DS math.STstat.ML ·2026-11-01 ·Jiale Cheng, Ziao Wang, Lei Ying
Abstract · 67 pages, 3 figures

This paper presents a near-quadratic-time algorithm that correctly matches almost all nodes of two correlated Erdős-Rényi graphs without any incorrect matches. Each graph has average degree $λ$ and its edges are subsampled from a common parent Erdős-Rényi graph with $n$ nodes, with subsampling probability $s$. The nodes of one graph are then relabeled by an unknown permutation. The algorithm is built on a test for correlated versus independent Galton-Watson trees with approximated likelihood ratios. For every constant $s>\sqrt{C_{\mathrm{Otter}}}$, where $C_{\mathrm{Otter}}\approx0.338$ is Otter's tree-counting constant, and every $λ=ω(1)$ with $λ=(\log n)^{O(1)}$, the algorithm outputs a partial matching of size $n-o(n)$ that contains no incorrect matches with high probability. If, in addition, $λs\ge(1+ε)\log n$ for a constant $ε>0$, a seeded completion step recovers the permutation exactly. In both cases, the time complexity of the algorithm is $n^{2+o(1)}$ with high probability, whereas previous polynomial-time algorithms with the same guarantees either have a time-complexity bound whose exponent goes to $\infty$ as $s$ approaches $\sqrt{C_{\mathrm{Otter}}}$ from above, or apply only when $s$ is close to $1$.

0.31

An approximate cactus-decomposition for vertex-connectivity

cs.DM math.CO ·2026-11-01 ·Romain Bourneuf, Johannes Carmesin, Jan Kurkofka, Tim Planken
theory
Abstract

The classical cactus theorem of Dinitz, Karzanov and Lomonosov from 1976 states that the minimum edge-cuts of a connected graph are displayed by a cactus. For vertex-separations, Tutte's theorem gives an exact cactus-decomposition for connectivity $2$, but exact cactus-decompositions fail for higher connectivity. Extending earlier approximate cactus-decomposition results for connectivities $3$ and $4$ by Carmesin and Kurkofka [FOCS 2023] and Kurkofka and Planken [SODA 2026] to all connectivities, we prove that for every $k\in\mathbb{N}$ there is a constant $d=d(k)$ such that every $k$-connected graph $G$ admits a cactus-decomposition displaying $k$-separations such that every $k$-separation of $G$ has edit-distance at most $d$ to a displayed separation. Our proof develops tools that may be useful beyond this result. The first is a three-dimensional version of the classical corner-diagram method, which we use to show that triples of crossing $k$-separations have a controlled two-dimensional behaviour. The second is a covering-space approach for studying local crossing patterns from a global perspective. Third, we advance the theory of flowers, which originated in matroid theory.

0.30

Polynomial-Time Minimax Prediction for Stationary Renewal Processes

cs.IT cs.LG ·2026-11-01 ·Xinyu Chen, Yiming Ma
theory
Abstract · 11 pages; includes exact algorithms and a reproducibility appendix with ancillary Python code

The minimax next-symbol Kullback-Leibler risk for stationary binary renewal processes is known to have order n^(-1/2). We give a polynomial-time predictor attaining this order for an arbitrary unknown finite-mean interarrival law. The construction uses a classical type-uniform renewal code. Its contribution is an exact dynamic program for censored prefix marginals, avoiding enumeration of gap histograms or binary completions. Combined with an established prediction reduction, it gives risk at most [log M_(n+1)+log(n+2)]/n, where M_N counts renewal types. Shared suffix updates reduce the total computation to O(n^3) arithmetic work and polynomial bit complexity, without tail, mixing, known-mean, or bounded-support assumptions. The statistical rate and its lower bound are prior results; the proposed advance concerns computation.

0.30

The Stability of Data Exchange in Competitive Markets

cs.GT ·2026-11-01 ·Yuanchen Brian Tang, Jiaxin Song, Bhaskar Ray Chaudhury, Ruta Mehta
Abstract

We study the stability of data exchange among competing firms. Unlike tangible goods, data is non-rival: sharing it does not deprive the giver of its use. On one hand, competing firms-- often operating as machine-learning-driven predictors that collect observations from a shared environment to solve similar tasks-- can improve their predictive accuracies from pooling data. On the other hand, they compete for market share, which may depend on their relative predictive performance. This raises the central question of this paper: can there be a stable data exchange in such competitive environments? We formalize this via a model of $n$ firms exchanging fractions of their datasets, with predictive performance feeding into market competition. We adopt core stability as our solution concept, accounting for externalities via worst-case responses by non-participating firms. Our main results reveal a sharp contrast driven by demand structure: (i) in fully rational markets, where customers flock entirely to the top-performing firm, a core-stable exchange always exists for any monotone continuous prediction functions; (ii) in partially rational markets with logit-style demand, core-stable exchanges may fail to exist-and even approximate cores can be empty. Our impossibility results suggest that internal profit-sharing within coalitions can restore stability in partially rational markets. We formalize this intuition by showing that when co-operating coalitions equally redistribute market-share revenues among members, a core-stable data exchange exists. Finally, we propose a natural exchange dynamic and demonstrate empirically that it converges to a core-stable data exchange in fully rational markets on simulated instances derived from real datasets.

0.30

A new class of optimal binary linear complementary pairs of codes

cs.IT ·2026-11-01 ·Yansheng Wu, Jing Chen
published
Abstract · 31 pages, accepted by SETA 2026

The construction of optimal binary linear complementary pair (LCP) has recently attracted considerable attention in coding theory. Based on the Belov codes, this paper proposes a novel construction of optimal binary LCPs with $θ=4$, extending our previous work on $θ=3$. Our results not only yield abundant new optimal binary LCPs with flexible parameters, but also enable more optimal binary LCPs with smaller minimum distances.

0.30

Factor Closure in Noncommutative Algebraic Complexity

cs.CC ·2026-11-01 ·V. Arvind, Pushkar S. Joglekar
theory
Abstract

We prove that $\VPnc$ is closed under taking factors over every field. If a noncommutative circuit of size $s$ computes a nonconstant polynomial of degree at most $D$, then each left or right factor has a circuit of size $O(sD^3)$. A central factor $g$ in a decomposition $f=agb$, with nonconstant outer factors $a,b$, has a circuit of size $O(sD^6)$. The proof uses a small weighted nondeterministic finite automaton (NFA) with replacement labels. Given $f=gh$, its labels depend on a word of maximum degree in $h$ and the coefficients of its suffixes. A pairing of accepting paths proves that the induced linear map $T$ satisfies $T(gh)=g$. Matrix substitution implements $T$ in a circuit. Applying the same substitution to a Boolean-sum verifier proves $\VNPnc$ closure. The construction also proves formula closure. The automaton has one copying state and, after leaving it, uses only scalar replacements. This structure gives formulas of size $O(s^2)$ for one-sided factors and $O(s^4)$ for central factors of a size-$s$ formula. The same map proves that polynomial-size noncommutative algebraic branching programs are closed under taking factors. Polynomial-size ABP factorizations over finite fields are already known from~\cite{AJ25}. These are arithmetic size bounds, with arbitrary field constants at unit cost, including over $\Q$. Over finite fields, combining the formula construction with~\cite{AJ22} gives randomized polynomial-time complete factorization of formula inputs, with factors output as formulas. For polynomial-degree circuit inputs, we obtain polynomial-length succinct ABP descriptions with polynomial-time edge queries. We also give two algorithms that output complete ordered irreducible factorizations as polynomial-size circuits: a deterministic $\FP^{\SigTwo}$ algorithm and a zero-error randomized algorithm with an NP oracle and expected polynomial running time.

0.29

Smoothed analysis of stability radii for the assignment problem

cs.DS math.OCmath.PR ·2026-11-01 ·Michael P. Friedlander
Abstract · 17 pages, 3 figures

Every linear program can tolerate small changes to its costs and still yield a solution of the original problem. The size of this tolerance measures how robust an optimal decision is to noise or numerical error, and how much regularization the problem can support. Classical sensitivity analysis can determine this tolerance only from an optimal solution. This paper provides a priori lower bounds on two measures of this tolerance -- the optimality gap and the $\infty$-norm stability radius -- for the $n$-by-$n$ linear assignment problem over the Birkhoff polytope. The bounds derive from a smoothed analysis, which adds Gaussian noise with standard deviation of at least $σ$ in every direction to an arbitrary base cost. At every fixed confidence level, uniformly over the base cost, both quantities are at least of order $σ/(n\sqrt{\log n})$. The radius bound is asymptotically sharp. Under isotropic noise, the worst case occurs when every assignment has equal cost, and in this case, a scaled stability radius converges in distribution to a standard exponential variable.

0.29

Concatenation Locality under Successive Adjacent Transpositions

cs.IT ·2026-11-01 ·Adelin Ataba, Aryeh Lev Zabokritskiy
theory
Abstract · 32 pages

We study codes correcting up to $t\geq1$ successive adjacent transpositions, using block concatenation to support blockwise encoding and decoding. A fixed set $P$ of words of length $b$ over a $q$-symbol alphabet generates, for every $m$, the length-$mb$ code of all concatenations of $m$ words from $P$. For our explicit constructions, the prepared encoder and decoder depend only on the block parameters and are reused at every concatenation length. We first prove that correction at every concatenation length is certified by a single finite test at $2t+1$ blocks, and this test is tight over a $q$-symbol alphabet exactly when $\binom q2\geq2t$. For binary words and $b\geq2t+1$, we give an explicit construction with total redundancy at most $mt\log_2 b+O_t(m)$, matching the optimal per-block redundancy $t\log_2 b+O_t(1)$ among block sets correcting at every concatenation length. Among all binary $t$-correcting codes of length $mb$, the optimal total redundancy is $t\log_2(mb)+O_t(1)$. Using all concatenations of one block set makes our total redundancy grow linearly in $m$ and buys independent block encoding, with one encoder and decoder at every length. The construction encodes and decodes in time linear in $mb$ for fixed $b,t$, and recovers the $j$-th source block from the first $jb$ received symbols. We also prove that, whenever the length-$mb$ code corrects $t$ transpositions, the $j$-th source block is determined within $2tb$ received symbols beyond position $jb$, and that explicit binary block sets with rates tending to one as $b$ grows with $t$ fixed attain this bound in the worst case. Finally, we prove that every ordinary correcting code contains a constant fraction of its words whose arbitrary concatenations retain correction, for fixed $q,t$.

0.28

On Guarding Line Arrangements

cs.CG math.CO ·2026-11-01 ·Valentin Brimkov
Abstract · 21 pages, no figures

We study the problem of guarding a set of straight lines in the plane by selecting a minimum-cardinality set of points such that every line contains at least one selected point. This problem is known to be strongly NP-hard \cite{megiddo}. We investigate approximation algorithms for the problem and analyze their performance guarantees under various structural assumptions on the underlying line arrangement. In particular, we present approximation schemes based on set-cover techniques, relaxation methods, and properties of maximal lines. We establish constant-factor bounds on the approximation ratio for several classes of arrangements and derive relationships between the guarding number and the incidence parameters of the arrangement. Connections with Dirac-type incidence results are also explored.

0.28

Integral saturation of adjacency-degree lattices

cs.DM math.CO ·2026-11-01 ·Zhipeng Lu
theory
Abstract

Let $A$ be the adjacency matrix of $G\sim G(n,1/2)$, let $D=\operatorname{diag}(A\mathbf{1})$, and let $Λ_G=\mathbb{Z}\langle A,D\rangle\mathbf{1}$. We prove $\Pr[Λ_G\ne\mathbb{Z}^n]\le n^{-3/20+o(1)}$. Thus two dependent graph operators generate the entire vertex lattice from the constant vector, simultaneously over every prime field. Adjacency alone has a modulo-two rank obstruction for every simple graph of order greater than one, even though its rational word span is typically full. The proof first constructs words containing at most two adjacency operations whose integer span contains $Δ_G^3\mathbb{Z}^n$, where $Δ_G$ is the product of the pairwise differences between distinct degree values. This confines all possible arithmetic obstructions to a finite prime range on a single event. We then remove the remaining obstructions using conditional alternating forms, symmetric random-matrix code estimates, and uniform counts of degree-graded invariant subspaces. We connect the integer lattice to graph reconstruction through pointed moments. Saturation forces every pointed orthogonal AD intertwiner to be a vertex permutation. The short-word construction also shows that a fixed polynomial-size family of ordinary caterpillars, of diameter at most seven and order $O(\sqrt{n\log n})$, identifies almost every graph among all finite simple graphs. For nonsaturated full-rank lattices, we give an exact description of the moment fibre by admissible unimodular overlattices; at index two, the only possible ambiguity is a specified four-vertex Godsil--McKay switch.

0.28

6G Uplink Waveform and Rank Selection under Distributed Reception

cs.IT eess.SP ·2026-11-01 ·Yewen Cao, Fen Hou, Yulin Shao
Abstract

The development of 6G uplink is expanding waveform and spatial transmission options, while distributed reception makes their relative merits depend on the spatial and frequency structure of the aggregate channel. We study the joint selection of cyclic prefix orthogonal frequency-division multiplexing and discrete Fourier transform spread OFDM at rank 1 and noncoherent rank 2 under common user equipment power constraints and centralized linear minimum mean-square error reception. We derive waveform switching thresholds at both ranks and a rank selection margin, establishing conditions for selecting among the four modes based on available transmit power, DFT despreading loss, and spatial multiplexing gain. We then characterize how port gains and correlations at individual access points (APs) form aggregate spatial separation and influence mode selection. At fixed port gains and available transmit powers, improving spatial separation without reducing it on any subcarrier increases the rank selection margin, while its effect on the preferred waveform within rank 2 depends on the frequency distribution of the improvement. Numerical results show that different AP deployments can favor different waveforms and ranks, even with the same receive antenna budget and comparable aggregate channel gains, including waveform switches at a fixed preferred rank.

0.28

Wideband Microwave Linear Analog Computer (MiLAC) for Frequency-Flat and Frequency-Selective Processing

cs.IT eess.SP ·2026-11-01 ·Juan Vidal Alegría, Matteo Nerini, Bruno Clerckx
Abstract · 15 pages, 6 figures. Copyright information may change upon publication at an IEEE journal

Microwave linear analog computer (MiLAC) has been proposed as an analog processing architecture to perform efficient computation in the microwave domain. Consisting of reconfigurable impedance networks, a MiLAC can extract many of the gains of large multi-antenna systems without costly digital processing, e.g., achieving capacity in narrowband multiple-input single-output (MISO). However, the behavior of wideband MiLAC, required to support high data-rate transmissions, is widely unknown. The impedance components of MiLACs have non-trivial frequency-dependent response, requiring specific implementations to characterize it. In this work, we consider MiLAC implementations based on LC resonators with tunable capacitance, which allow synthesizing near-arbitrary passive lossless impedance matrices at individual frequencies using inductors and varactors. We characterize the mapping between optimal MISO precoding at a reference frequency, and the vector of capacitances configuring the MiLAC. We derive the degrees of freedom (DoFs) remaining after fixing such optimum precoding at the reference frequency, and show that these DoFs can be exploited to tune the frequency response of the MiLAC. We propose two techniques to exploit these DoFs: a closed-form approach for frequency-flat MiLAC processing, and a numerical approach for multi-frequency MiLAC processing, optimizing processing at frequencies other than the reference one. The frequency-flat approach shows satisfactory performance in a simulated frequency-flat millimeter-wave scenario, approaching the performance of digital processing. The multi-frequency approach can adapt to the frequency-selectivity of a simulated sub-6GHz scenario, attaining gains over frequency-flat digital processing. This constitutes the first demonstration that realistic MiLAC implementations can be configured to perform independent computations at different frequencies.

0.27

Three-phase time minimization transportation problem with non-idle phases: a polynomial two-witness algorithm

cs.DS math.OC ·2026-11-01 ·Sailesh Kamalahasan, Raghav Paladugu, Archana Khurana
theory
Abstract · 44 pages (37-page main article and 7-page supplement), 0 figures; reproducibility code and data included as ancillary files

The three-phase time minimization transportation problem partitions the origin-terminal links into three sets used in consecutive phases and minimizes the sum of the phase shipment times. Optimal solutions may leave a phase idle, as in the main example of the original study. We require every phase to carry a positive integer quantity while leaving the split of the total requirement across phases free. Discarding idle solutions from the original algorithm does not, in general, solve this variant. We prove that the optimum is attained by fixing one witness link in each of two phases and deciding the third phase by a single network flow computation, with the optimal Phase-3 time of each witness pair located by binary search. The resulting exact algorithm is strongly polynomial. It extends to any fixed number of phases, whereas, when the number of phases is part of the input, even non-idle feasibility is strongly NP-complete; feasibility alone is fixed-parameter tractable in the number of phases via known rainbow-matching results. We also characterize the minimum time of each phase. On the main example, non-idleness raises the optimum from 38 to 41. On 3,650 random instances with up to $50\times50$ links the algorithm took at most 50.2 seconds, and non-idleness raised the optimal value in 41% of the 2,250 instances with equal phase sizes; on 90 sampled instances it was faster than a mixed integer programming solver on 80 with a direct formulation and on 65 with a tighter threshold formulation.

0.27

Instance-Specific Sharp Convergence Rate of the Add-$λ$ Estimator via Feature Space Decomposition

cs.IT math.ST ·2026-11-01 ·Zong Shang
Abstract

Using the Feature Space Decomposition framework, we establish non-asymptotic matching upper and lower bounds for the expected Kullback-Leibler estimation error of the Add-$λ$ estimator of a discrete probability distribution, for any signal, any dimension, any sample size, and any $λ>0$.

0.26

Monotone Local Search with Sample-Dependent Running Times

cs.DS cs.DM ·2026-11-01 ·Serge Gaspers
theory
Abstract · Companion site (including certificates and Lean formalization): https://serggasp.github.io/sample-MLS-site/ ; archived at https://doi.org/10…

Monotone Local Search turns an algorithm that extends a given partial solution into an algorithm that finds a solution from scratch: sample a few elements, hope that they lie in an unknown solution, and complete them. Its known analyses bound every call of the extension algorithm by a worst-case time that depends only on the number of sampled elements, and Esmer et al. (SODA 2024) showed that no algorithm that uses the extension algorithm as a black box with this running time beats the original form by more than a polynomial factor. We let this running time depend on the actual sample. If the extension algorithm decides in time $O^*(2^{g(I,X,k)})$, for a polynomial-time computable $g$, whether a sample $X$ can be completed to a solution with at most $k$ elements, our main theorem gives a randomized algorithm with one-sided error in time $O^*(2^{ψn})$ whenever, for some solution $S$ of size $k$ and some $t$, the exponential moment $\sum_{X\subseteq S,|X|=t}2^{-g(I,X,k)}$ is at least $2^{-ψn}\binom{n}{t}$. A corollary, multivariate Monotone Local Search, generalizes the theorems of Fomin et al. (STOC 2016, J. ACM 2019) and of Gaspers and Lee (ICALP 2017). For Max Leaf Spanning Tree and, equivalently, Connected Dominating Set, the previous best polynomial-space bound is $O^*(1.8619^n)$. Completing a sample is a Steiner tree problem, solved in time exponential in the number of terminals, which varies strongly with the sample. We obtain a randomized polynomial-space algorithm in time $O(1.6991^n)$ and a deterministic one in time $(\sqrt5-\frac12)^n 2^{o(n)}=O(1.7361^n)$. For connectivity problems in which the extension algorithm's running time doubles with every component of the sample, an analysis through a partition function gives randomized polynomial-space algorithms in time $O^*(1.8^n)$ for Connected Feedback Vertex Set and $O(1.6444^n)$ for Connected Vertex Cover.

0.26

A logarithm-free subexponential algorithm for minimum quartet inconsistency

cs.DS ·2026-11-01 ·Chuang-Chieh Lin
theory
Abstract

We give a deterministic algorithm for complete, unweighted minimum quartet inconsistency (MQI) with time and space $2^{O(k^{1/4})}+n^{O(1)}$, where $n$ is the number of taxa and~$k$ is the allowed number of quartet disagreements. This removes the logarithmic factor from the exponent of the MQI bound announced in 2009. The algorithm combines low-error anchoring, chromatic coding, and dynamic programming that accounts for disagreements with all input quartets, with a refined count of colorings and merging states. We also record a generic ETH-based lower bound. The optimality of the fourth-root exponent remains open.

0.25

The Runtime-Moment Frontier for Universal Sampling

cs.CC ·2026-11-01 ·Sebastian Ben Daniel
Abstract

Can one sampler give every string at least a polynomial fraction of the probability that any source in a class gives it, while keeping a bounded moment of its own runtime? We answer this in terms of runtime moments. For every oracle and every rational $a>0$, one sampler with polynomial $a$th runtime moment dominates every source with polynomial $r$th moment with power $B(a,r)=1+(a-\min\{r,1\})_+/r$, and the Rényi divergence of order $B/(B-1)$ of the source from it is $O(\log n)$, a sharp quantitative form of Gurevich's domination on average. Relative to an oracle, $B(a,r)$ is optimal in both senses, with failures at every large input length. So the first moment is a threshold: one sampler of order one is universal for expected polynomial time, but none of higher order is universal even for polynomial-space sources with all polynomial moments. For expected-polynomial-time sources the threshold also holds relative to a random oracle and, without an oracle, under a sequential-hardness hypothesis on iterated functions or a parallel-sequentiality hypothesis on verifiable delay functions with short seeds. The frontier governs algorithms as well. For rational orders we determine the exact exponent by which polynomial runtime moments of deterministic deciders transfer from one fixed sampler to every source, and show, relative to oracles, that all correct solvers of a sparse language in $\mathsf{UP}\cap\mathsf{coUP}$ meet it. Where $B(a,r)=1$, actual running times transfer better than arbitrary costs, but only above a critical moment.

0.25

Efficient Lempel-Ziv Prediction with the Run-Length BWT

cs.DS cs.LG ·2026-11-01 ·Ohad Avnery
Abstract · 21 pages, 1 table

Lempel-Ziv Plurality (LZP) is an algorithm of Kosoy that predicts the next symbol of a string by using its Lempel-Ziv factorization, and makes at most $O(z\log n)$ mistakes on every string of length $n$ with $z$ LZ77 phrases. We give efficient online implementations of LZP and of a new variant of it. Our first implementation maintains a run-length encoded Burrows-Wheeler transform (RLBWT) of the reversed history. It implements LZP, uses $O(z\log z\log^2 n)$ bits, and has worst-case update and prediction times $O(\log z+\log\log n)$ and $O(σ(\log z+\log\log n))$, respectively, where $σ$ is the size of the alphabet. This is an exponential improvement over Kosoy's implementation, whose update time is $O(z\log^7 n)$. We then introduce the Lempel-Ziv Majority (LZM) predictor, which has the same mistake bound and update time, and predicts in $O(1)$ time independently of $σ$. These two predictors accept new symbols at any time, without knowing the alphabet in advance. For an alphabet known in advance, a different data structure gives $O(σ(1+\log z/\log\log n))$ time per symbol, which is constant when $σ=O(1)$ and $z=\mathrm{polylog}(n)$.

0.25

Multiplicative Weights for Mixed Block Constraints and Maximum Multicommodity Flow

cs.DS ·2026-11-01 ·Vladimir Kolmogorov
Abstract · 15 pages

We study multiplicative-weights (MWU) algorithms for block-separable feasibility problems containing both packing and covering constraints. Each iteration optimizes a weighted linear objective over one block, either chosen uniformly at random or visited in a fixed cyclic order. We show that the cyclic algorithm returns an $O(\varepsilon)$-approximate solution after $O(kρ\varepsilon^{-2}\log m)$ oracle calls where $k$ is the number of blocks and $ρ$ is the width of the oracle. The same holds for the randomized version with high probability. Previously, such bounds were shown when the coupling constraints have the same sign [Radzik'95; Villavicencio-Grigoriadis'96]. These analyses critically depended on monotonicity arguments that are not available in the mixed case. As an application, we obtain a deterministic $(1-\varepsilon)$-approximation for maximum multicommodity flow (M-MCF) using $O(k\varepsilon^{-2}\log m\log(km/\varepsilon))$ approximate single-commodity min-cost-flow calls. For planar graphs, and more generally for graph families admitting a near-linear-time min-cost-flow algorithm, this gives total running time $\widetilde O(km\varepsilon^{-2})$. Compared with the $O(km^{1+o(1)}\varepsilon^{-1})$ algorithm of [Chen-Graur-Sidford'25], this improves the dependence on $m$ for these graph families at the cost of a worse dependence on $\varepsilon$. Finally, we give an exponentiation-free finite-precision implementation of the cyclic algorithm. It represents each internal weight by a logarithmic-bit mantissa and exponent, while every price supplied to the block oracle is rational and has a polynomially bounded common denominator.

0.24

Deterministic $n^{O(\log n)}$ Graph Isomorphism Algorithm via Inherited Decompositions

cs.DS ·2026-11-01 ·MohammadTaghi Hajiaghayi, Mahdi JafariRaviz
Abstract

We give a deterministic $n^{O(\log d)}$-time algorithm for isomorphism of graphs with $n$ vertices and maximum degree $d$. Since $d<n$, this is an $n^{O(\log n)}$-time algorithm for graph isomorphism. To our knowledge, this is the first proof of such a worst-case bound for general graphs. It improves the explicit cubic logarithmic exponent in the analysis of the corrected Luks--Babai algorithm. For graphs of maximum degree $d$, the best previous bound known to us, due to Grohe, Neuen, and Schweitzer, is $n^{O((\log d)^c)}$ for an absolute constant $c$. The algorithm uses the local-certificate and color-refinement steps of Babai's quasipolynomial algorithm, but retains the decomposition of each recursive instance when it solves an auxiliary coloring problem. There are two reasons for retaining this information. First, a local test can make progress without removing many vertices: it can eliminate a layer of permutation symmetry. A logarithmic bound on the total weight of these layers allows tests of constant size. Second, the auxiliary bipartite graphs arising in the symmetry-breaking step can be represented inside the current instance, after retaining a row decomposition when necessary. These recursive calls inherit the progress already made. Two further observations give the dependence on the degree. The recursive calls made by a local test check disjoint parts of the instance, so together they cost about as much as one call on the whole instance. And for graphs of maximum degree $d$, each large symmetric action that the algorithm must break acts on at most $d$ points. We present the algorithm through colorings, partitions, and bipartite graphs, and collect the permutation-group ingredients in one appendix. The result extends to string isomorphism and returns one isomorphism and generators for the full automorphism group.

0.23

Sharper Bounds for Graphical Two-Choice Allocation

cs.DS math.PR ·2026-11-01 ·Obinna Okechukwu
benchmarktheory
Abstract · 28 pages, 5 figures

In graphical two-choice allocation, vertices are bins, and each arriving ball must be placed irrevocably at one endpoint of a uniformly random edge. The objective is to keep the maximum-minus-minimum load gap small at every time. On cycles, endpoint-local monotone rules incur polynomial gaps at long time horizons, whereas global rules can achieve polylogarithmic gaps. We prove a joint exponential-moment inequality for the orthogonal contrasts of a binary partition tree, controlling all contrasts along a root-to-leaf path simultaneously. Applied to the Bansal--Feldheim strategy, it improves the gap on every $κ$-edge-connected $d$-regular graph from $O((d/κ)\log^4 n\log\log n)$ to $O((d/κ)\log^3 n\log\log n)$. We give an explicit strategy on the $n$-cycle with expected gap $O(\log^2 n)$ uniformly over deterministic times and exponential tails at each such time. An Euler-tour projection yields $O(\bar d\log^2(2M))$ for every connected graph with $M$ edges and degree-proportional capacities. For lower bounds, we introduce a stochastic cut workload: the expected largest normalized overload forced by the requests, plus an indivisibility term. For every online integral strategy, the supremum over time of the expected gap is at least one sixth of this benchmark. Regular rings of cliques attain $Ω((d/κ)\log n)$ along an explicit family, improving the prior additive lower bound. With unequal capacities, the workload need not control the gap: one family has constant relative cut slack, bounded workload and gap $Ω(\log\log n)$; another has workload at most $20$ but forces gap $Ω(\log n)$ even for clairvoyant fractional allocations that remain irrevocable.

0.22

Secrecy Sum-Rate Maximization for RSMA-Enabled Pinching-Antenna Systems

cs.IT eess.SP ·2026-11-01 ·Code Kebe, Saeid Pakravan, Wessam Ajib
Abstract

This paper investigates secrecy sum-rate (SSR) maximization in a downlink rate-splitting multiple access (RSMA)-enabled single-waveguide pinching-antenna system (PASS) with an external eavesdropper. The RSMA power allocation, secure common-rate allocation, and pinching-antenna (PA) positions are jointly optimized subject to transmit-power, per-user secrecy-rate, and PA-placement constraints. To tackle the resulting nonconvex problem, an alternating-optimization framework is developed, combining successive convex approximation for RSMA resource allocation with feasibility-preserving projected gradient ascent for PA-position optimization. Numerical results show that the proposed design consistently outperforms the considered benchmarks in terms of average SSR, demonstrating the benefits of joint RSMA resource and PA-position optimization for secure PASS transmission.

0.22

Communication with Adversarial Users: Byzantine Models in Multi-User Information Theory

cs.IT ·2026-11-01 ·Minglai Cai, Christian Deppe, Neha Sangwan
survey
Abstract

We consider communication systems operating in environments where some participating users cannot be trusted. This motivates the study of Byzantine communication models, in which users may behave adversarially. In this survey, we provide a unified information-theoretic perspective on communication with adversarial users in multi-user settings. We review recent advances in Byzantine multiple access channels, including reliable communication, adversary identification, and distributed function computation. Furthermore, we discuss extensions to classical-quantum channels and highlight fundamental differences between classical and quantum models. Finally, we outline key open problems and future research directions at the intersection of communication, computation, and security.

0.21

A Deterministic O(log m)-Competitive Algorithm for Online Minimum Cost Bipartite Matching with Delays

cs.DS ·2026-11-01 ·Tung-Wei Kuo
theory
Abstract

In the online Minimum cost Bipartite Perfect Matching with Delays (MBPMD) problem, $m$ requests and $m$ servers arrive over time in a metric space. An online algorithm has to match each request to a server and can postpone a matching decision by paying the delay cost. The objective is to minimize the total distance and delay cost. Azar, Chiplunkar, and Kaplan gave a randomized $O(\log n)$-competitive algorithm for metric spaces with $n$ points that are known in advance. For deterministic algorithms, however, the best known competitive ratio in general metric spaces was $O(m^{0.59})$, due to Azar and Jacob-Fanani. On the line, Kuo proposed the Virtual Robust Matching (VRM) algorithm and showed that it is $O(\sqrt{m}\log^2 m)$-competitive. We prove that VRM is $O(\log m)$-competitive for the MBPMD problem in any metric space. VRM is deterministic and does not need to know the metric space or $m$ in advance. It is the first deterministic polylogarithmic-competitive algorithm for the MBPMD problem in general metric spaces. Unlike the previous analysis of VRM, our analysis does not rely on a geometric decomposition of the metric space. Instead, it charges the waiting time of requests over time to the edges of a fixed optimal matching. We also extend VRM to the non-bipartite version and prove that the resulting deterministic algorithm is $O(\log m)$-competitive. This improves the $O(\log^5 m)$ bound of Dufay and Wattenhofer and matches a concurrent result of Ahmadian, Chawla, Kumar, Purohit, and Zhang. The proof that VRM is $O(\log m)$-competitive and its non-bipartite extension were found by GPT-6 Astra (OpenAI) with almost no human input. We used Claude Opus 5.5 (Anthropic) to simplify and present the results. See the AI Usage and Methodology section for details. We take full responsibility for this paper.

0.20

Faster Algorithms for Disjoint Paths in DAGs and Two Disjoint Shortest Paths

cs.DS ·2026-11-01 ·Yang Cao, Haoran Qi, Hanzhi Wang
Abstract

For $k$ vertex-disjoint paths with prescribed endpoints in a directed acyclic graph (DAG), the $O(mn^{k-1})$-time pebbling algorithm of Fortune, Hopcroft, and Wyllie [TCS 1980] has remained the fastest for every $k\geq3$ for over four decades, even when the paths must be shortest. We break this bound for every $k\geq3$ with randomized algorithms running in expected $O_k(m+n^{k-3+ω+\varepsilon})$ time for every $\varepsilon>0$, where $ω<2.372$ is the matrix multiplication exponent; this is $O_k(m+n^{k-0.62})$. For two disjoint shortest paths, we find the paths as fast as we detect them: in expected linear time on DAGs and on undirected graphs with polynomially bounded integer weights, and in expected $O(\min\{mn,n^{ω+\varepsilon}\})$ time on directed graphs whose cycles have positive weight, which on dense graphs is $o(mn)$ even for detection. All algorithms evaluate a polynomial whose monomials are the solutions: a new first-collision circuit speeds up the evaluation in DAGs, and a correlated extraction step at cuts crossed once by every path finds a solution at the cost of a constant number of evaluations.

0.20

Exact Tie-Breaking and Tight Degree Bounds for Minimum-Degree Greedy on Interval Graphs

cs.DS ·2026-11-01 ·Ruize Zhao
Abstract · 23 pages, 3 figures

We study tie-breaking in the minimum-degree greedy algorithm (MDG) for maximum independent set on interval graphs. Given an interval representation of an $n$-vertex graph, we compute the largest and smallest output sizes and construct executions attaining them in $O(n^3)$ time and $O(n^2)$ space. For maximum degree at most four, favorable tie-breaking gives a tight guarantee of $3/4$. For every maximum-degree bound of at least five, the guarantee drops to $2/3$. Both bounds are tight even on connected unit interval graphs whose output size is independent of tie-breaking. Together with the known optimality of favorable tie-breaking for maximum degree at most three, these results give a complete classification by degree bound.

0.17

Faster graphic matroid parity: Breaking the $nm$ time complexity barrier

cs.DS ·2026-11-01 ·Tatsuya Terao
Abstract · 51 pages, 1 figure. The main algorithmic idea is due to the author. Using this idea, the author developed a $\tilde{O}(n^{4/3}m^{2/3}+\sqrt{…

In the graphic matroid parity problem, we are given an undirected multigraph $G=(V,E)$ with $n$ vertices, together with a partition of $E$ into $m$ pairs of distinct edges, and the goal is to find a maximum-cardinality collection of pairs whose union is a forest. This problem generalizes both graph matching and graphic matroid intersection. Graph matching admits the celebrated $\tilde{O}(\sqrt{n}m)$-time algorithm of Micali--Vazirani~[FOCS'80], and graphic matroid intersection admits $o(nm)$-time algorithms due to Gabow--Stallmann~[ICALP'85] and Gabow--Xu~[FOCS'89]. In contrast, for graphic matroid parity, Gabow--Stallmann~[FOCS'84 \& Combinatorica'86] gave an $O(n^2 m)$-time algorithm, and its $\tilde{O}(nm)$-time implementation by Gabow--Stallmann~[ICALP'85] had long remained the best known bound, except in very dense regimes where $m=ω(n^3)$, in which case the algebraic $\tilde{O}(n^4)$-time algorithm of Cheung--Lau--Leung~[SODA'11 \& TALG'14] is faster. We present a randomized $\tilde{O}(n^{3/2} \sqrt{m} + \sqrt{n} m)$-time algorithm for graphic matroid parity. This is the first algorithm to break the $O(nm)$-time complexity barrier for this problem outside the very dense regime. Moreover, it improves over the previous bounds except in very sparse regimes where $m=\tilde{O}(n)$ and in very dense regimes where $m = \tildeΩ(n^{3.5})$. In particular, when $m=Θ(n^2)$, our algorithm runs in $\tilde{O}(n^{2.5})$ time, which nearly matches the $\tilde{O}(n^{2.5})$-time bounds of Micali--Vazirani for graph matching and of Gabow--Xu for graphic matroid intersection in this regime.

0.16

Faster Semi-Streaming and Dynamic Approximate Maximum Weight Matchings in General Graphs: A Unified Framework

cs.DS ·2026-11-01 ·Jiale Chen
Abstract · Abstract shortened due to arXiv requirements

We study the $(1-\varepsilon)$-approximate maximum weight matching problem in general graphs with $n$ vertices and integer edge weights in $[1,W]$. For $W=n^{O(1)}$, we obtain faster algorithms in the semi-streaming and dynamic settings: (1) Semi-streaming: Deterministic $O(\varepsilon^{-2})$-pass and $O(\varepsilon^{-2}\log(1/\varepsilon))$-pass algorithms for unweighted and weighted matching. (2) Incremental: A deterministic algorithm for weighted matching processing $m$ edge insertions in total time $O((m\varepsilon^{-3}+n\varepsilon^{-6})\log^3 n)$. (3) Decremental: A randomized algorithm for weighted matching processing all edge deletions from an $m$-edge graph in total time $O((m\varepsilon^{-8}+n\varepsilon^{-47})\log^{O(1)}n)$ with high probability. (4) Fully dynamic: A deterministic reduction from weighted matching in general graphs to unweighted matching in bipartite graphs incurring $O(\varepsilon^{-4}\log^{O(1)}(1/\varepsilon))$ multiplicative update-time overhead, independent of $n$. Underlying these results is a unified framework that reduces the problem to repeatedly solving two simpler matching subproblems: a $(1-\varepsilon)$-approximate maximum weight matching on a sparse subgraph with $O(n/\varepsilon)$ edges, and a constant approximation on the remaining edges. The framework builds on dual entropy regularization, a smoothing technique underlying efficient solvers for packing linear programs. To handle the exponentially many odd-set constraints, we adapt the smoothing technique to regularize only the vertex duals and develop a new cut partition that preserves odd-set feasibility during optimization.

0.16

Quantum Lower Bounds for Collision Problems

cs.CC quant-ph ·2026-11-01 ·Pan Peng, Jingyu Wu
theory
Abstract

We establish tight lower bounds on the adaptive and non-adaptive quantum query complexities of Promised $k$-to-one Finding on an exactly $k$-to-one input string $x\in\mathbb Z_q^n$, for every fixed $k\ge2$ and every feasible alphabet size $q\ge n/k$. The lower bounds are $Ω_k(n^{1/2-1/(2(2^k-1))})$ for unrestricted adaptive algorithms and $Ω_k(n^{1-1/k})$ for one-round non-adaptive algorithms. The former matches the adaptive quantum algorithm of Liu and Zhandry [LZ19], whereas the latter matches classical birthday sampling even when the parallel queries are arbitrarily entangled. We also establish the non-adaptive lower bound $Ω_k(\sqrt{n/k})$ for $k$-to-one Testing when $q\ge n$, resolving an open question posed by Koiran, Landes, Portier, and Yao [KLPY10]. Our proofs generate the balanced input from a uniformly random permutation and use compressed permutation oracles to describe the information learned by the algorithm. The adaptive proof bounds the growth of partial collisions through successive queries. The one-round proof instead analyzes the state after the parallel queries directly, bounding the probability that the compressed database contains a full collision. In both proofs, a short verification step accounts for correct outputs that include unrecorded positions. For the decision problem, we also compare one-round states with independent and distinct output labels.

0.13

An $\mathsf{FPRAS}$ for Mixed Volumes

cs.DS cs.CCcs.CGmath.CO+1 ·2026-11-01 ·Xiaoyu Chen, Kuikui Liu
Abstract

We design the first fully polynomial-time randomized approximation scheme ($\mathsf{FPRAS}$) to estimate the mixed volume of $n$ arbitrary compact convex sets in $\mathbb{R}^{n}$ specified by weak membership oracles. This disproves a longstanding conjecture of Gurvits (2009). $\textbf{AI Disclosure}$ The main ideas were generated entirely by GPT 6 Astra. The purpose of this paper is to provide a human-written exposition of the algorithm.

0.12

Fresh-Priority Eviction for Cuckoo Filters: Exploiting Latent Randomness Under XOR Hashing

cs.DS ·2026-11-01 ·Min Liu
theory
Abstract

Cuckoo filters are a widely adopted approximate membership query (AMQ) data structure that supports insertions, lookups, and deletions with low memory overhead. At high load factors, however, insertion performance degrades sharply due to the random-walk eviction process, which may require a large number of element relocations. Recent work on bucketized cuckoo hash tables has shown that \emph{fresh-priority eviction}, which preferentially evicts elements whose alternate bucket location has not yet been used, achieves $O(\ell m)$ total expected evictions for bucket size~$\ell$, an improvement over the standard random walk. In this paper, we show that this technique transfers to \emph{cuckoo filters} despite their use of XOR-based partial-key hashing. The key insight is that a fresh element's fingerprint, although stored in the table, is not examined by the eviction policy, so when the fingerprint length satisfies $k = Ω(\log m)$, the alternate bucket hash remains a near-uniform random variable, a property we call \emph{latent randomness}. We prove that under this condition the corruption graph remains subcritical and the total number of evictions is $O(\ell m \log m)$; we conjecture the tighter $O(\ell m)$ bound holds via the full amortized analysis of Bell and Kuszmaul. We further propose encoding the fresh/kicked state implicitly by the element's position within its bucket, requiring no additional per-element metadata. Experimental results on a filter with $2^{22}$ slots and $8$-bit fingerprints demonstrate $1.2\times$ to $2.4\times$ fewer evictions at load factors between $94\%$ and $99\%$ for bucket size $\ell = 8$, and a $1.73\times$ reduction in cumulative evictions to full capacity.

0.10

MDS Algebraic Geometry Codes of Square-Root Length on Hyperelliptic Curves

cs.IT math.AG ·2026-11-01 ·Chuangqiang Hu, Guorui Ma
Abstract · 10 pages, 2 figures

For every fixed genus $g\geq2$ and every $0<ρ<1$, we construct maximum distance separable (MDS) algebraic geometry codes over $\F_q$ for all sufficiently large odd squares $q$. Their parameters are $\bigl[n,\lfloorρn\rfloor,n-\lfloorρn\rfloor+1\bigr]_q$ with $n=\sqrt q+1+O_g(q^{1/4})$. The codes use full Riemann--Roch spaces on genus-$g$ hyperelliptic curves. They are not monomially equivalent to generalized Reed--Solomon (GRS) codes when $q$ is sufficiently large. An infinite subfamily has exact length $n=\sqrt q+2gq^{1/4}+1$.

0.09

Robotics, Control & Signals 50

Assessing Temporal Dispersion of Time-Varying Channel at 38 GHz for Different Reference Signals

eess.SP cs.IT ·2026-11-01 ·Jarosław Wojtuń, Bartosz Czaja, Tomasz Grasza, Rafał Przesmycki, Jan M. Kelner +7 more
Abstract · 5 pages, 7 figures

This paper investigates the influence of reference waveform selection on temporal dispersion estimation in a time-varying radio channel at 38 GHz. Three reference waveforms-numerically generated in MATLAB, measured through a conducted connection, and recorded in an anechoic chamber -are compared using the same wideband channel measurements conducted in a dynamic road environment. Their effects on the estimated power delay profiles (PDPs), mean delay, and root mean square (RMS) delay spread are evaluated. Despite noticeable differences in the time-domain and spectral characteristics of the reference waveforms, the resulting PDPs exhibit similar multipath structures, while the estimated temporal dispersion parameters show relatively small differences. The results indicate limited sensitivity to reference waveform selection under the investigated measurement conditions.

0.49

Parameterisation of NMC and LFP 4680 cells for electrochemical and thermal modelling

eess.SY cs.CEphysics.chem-ph ·2026-11-01 ·Remus Teodorescu, Nicolai A. Weinreich, Berke Bilgin, Yi Zhuang, Wenzel Prochazka
code/dataunusual
Abstract · 36 pages, 6 figures, 4 tables; supplementary material included. Submitted to Frontiers in Batteries and Electrochemistry. Data: https://gith…

An electrochemical-thermal model of the 4680 cylindrical cell is presented: a Doyle-Fuller-Newman model of the electrode stack coupled to a two-node thermal balance of the roll and the can (a thermal DFN, TDFN), solved by an implicit finite-volume scheme. Two cells are parameterised from teardown geometry and open literature, with every value tagged by provenance: the Tesla NMC811/graphite tabless cell and the BYD FC4680 LiFePO4 multi-tab cell. The reaction heat is evaluated at the open-circuit potential of the particle-mean concentration, which restores the heat of mixing. Both sets are calibrated in the same way, with the test fixture identified from the measurement alone. Against the 1C and 2C charge test of Ank et al., the calibrated NMC set follows the mean can temperature within 0.52 and 1.46 K, with a contact heat of 0.94 mOhm matching the authors' estimate for their holder. For the LFP cell, 1C and 2C discharges to 2.0 V were measured on the authors' bench; the fixture conductance (0.337 W/K) follows from the energy balance and the heat capacity (277 J/K) from the cooling. With the end-of-lithiation drop of LiFePO4 restored in the cathode open-circuit potential, the calibrated set follows the can temperature within 0.44 K (1C) and 0.76 K (2C), reproduces the peaks and the relaxation after cut-off. What remains is what a single can node and a single representative particle do not describe: the axial gradient of the NMC can under contact heating, and a gradual decline of the LFP voltage before the knee at 2C, which points to the distribution of particle states in the electrode. The parameter sets (BPX, CSV) and the bench data are openly available.

0.47

A Neuroprosthetic SoC Featuring Rapid Local Sensory Feedback with On-Chip Unsupervised Multi-Class Spike Sorting

eess.SP ·2026-11-01 ·Bowen Liu, Yuhan Hou, Yu Huang, Jianxiong Xu, Hao You +8 more
publishedbenchmark
Abstract · 4 pages, 9 figures, 3 tables. Author's accepted manuscript. Accepted for publication in IEEE Solid-State Circuits Letters

This letter presents a 65 nm neuroprosthetic system-on-chip (SoC) featuring a rapid feedback pathway from motor recording to sensory stimulation. To enable low-latency sensory mapping modulated by individual-neuron firing rates, the SoC integrates a mixed-signal unsupervised spike-sorting module with stimulation-artifact rejection, a programmable sliding-window firing-rate calculator, and a closed-loop controller for adapting stimulation parameters. The spike-sorting module combines event-driven analog feature extraction with Manhattan-distance-based digital clustering while consuming 1.6 $μ$W/channel. On the Quiroga Difficult-1 benchmark, it achieves 94.58% three-class accuracy, with only a 0.4% reduction when stimulation artifacts are introduced at 20% of the spike count. The closed-loop controller supports polynomial and logarithmic mapping functions and incorporates Mitchell's approximation for power saving. The SoC is fully characterized on the bench and in vivo.

0.47

Distributed State Estimation over Wireless MIMO Fading Channels with Imperfect CSI

eess.SY ·2026-11-01 ·Xinming Yu, Chan Wang, Ming-min Zhao, An Liu, Songfu Cai
Abstract · 27 pages

We consider distributed state estimation of a linear plant under stabilizing feedback over wireless multiple-input multiple-output (MIMO) fading channels, where channel-estimation noise corrupts the channel state information (CSI) used for information fusion. We propose a distributed estimator that combines local prediction, communication over wireless MIMO channels, and a posterior update based on local information and communicated data. Each sensor uses noisy pilots to estimate its aggregated wireless MIMO fading channel, and a pilot-noise bias correction recovers the required neighboring-data differences in conditional mean. We first analyze the mean estimation error and show that a Chebyshev design of the communication weights achieves a prescribed decay factor $d\in(0,1)$ satisfying the sensing conditions, with finitely many communication updates per plant sample at every finite signal-to-noise ratio (SNR) gain. We then characterize mean-square stability and the effects of SNR gain and receive-array size on the state estimation mean-square error (MSE), including its transient and steady-state value. In the ideal-communication limit, we show that the square root of the absolute MSE deviation from its steady-state value decays as $o(d^k)$. Numerical simulations verify the theoretical results and show substantial performance gain compared to various state-of-the-art baselines.

0.39

Leveraging high-performance head-only gradients to enable ~2-fold higher multiband acceleration in SMS-EPI and SMS-EPTI

eess.SP ·2026-11-01 ·Yimeng Lin, Zachary Shah, Mehdi Khalighi, Zimu Huo, Mengze Gao +12 more
Abstract · 8 pages, 5 figures

High-performance head-only gradient systems can substantially reduce echo spacing (ESP) in echo-planar imaging (EPI), but the resulting encoding speed is typically used to shorten readouts and reduce distortion. Here, we instead leverage the increased echo budget to enable denser k-space sampling and higher simultaneous-multislice (SMS) acceleration in SMS-EPI and SMS-EPTI. Fast k-space traversal with denser ky-kz sampling is achieved using peripheral nerve stimulation (PNS)-optimized gradient waveforms, while high-quality reconstruction is obtained using field-correcting GRAPPA (FCG). The resulting ESPs on MAGNUS 3T and Impulse 7T are approximately half those of high-performance whole-body systems, enabling approximately 2-fold higher multiband (MB) acceleration. This facilitates high-quality 2.5-mm isotropic whole-brain resting-state fMRI at 7T with MB28 and a repetition time (TR) of 123 ms.

0.39

Back to the Future: Predicting Long-Horizon Dynamics with Retrospective Correction for Disturbance-Aware Retrieval

cs.RO cs.LG ·2026-11-01 ·Jiwoong Choi, Haoyu Li, Yiwei Lyu
Abstract · 8 pages, 8 figures, 1 table. Submitted to ICRA 2027

Retrieving objects from densely packed storage requires anticipating how contact disturbs surrounding objects when limited clearance prevents lifting the target over clutter. However, contact can propagate beyond the target's immediate neighbors, while accumulated prediction errors make learned dynamics unreliable over long planning horizons. We introduce DARE (Disturbance-Aware REtrieval), a geometry-aware relational dynamics framework that uses retrospective correction to improve long-horizon prediction for disturbance-aware retrieval. DARE encodes compact object descriptors and predicts action-conditioned disturbance propagation through object-level message passing. When a sparse observation becomes available, a learned corrector uses the prediction discrepancy to amend an earlier predicted state and reruns the intervening dynamics, propagating the correction through the predicted contact chain. These predictions guide a receding-horizon planner that balances retrieval progress against displacement and rotation of surrounding objects, permitting necessary contact while discouraging harmful disturbance. Experiments in planar and RoboCasa environments evaluate long-horizon prediction, generalization, and robustness for disturbance-aware retrieval in dense clutter. See our project website: https://the-back-to-the-future.github.io

0.38

Ray-Tracing-Parameter-Based Lightweight Hybrid Precoding via Cascade XGBoost for Wideband Massive MIMO-OFDM Systems

eess.SP cs.IT ·2026-11-01 ·Po-Heng Chou, Wei-Chen Lin, Wan-Jen Huang, Walid Saad, Ronald Y. Chang
Abstract · 6 pages, 4 figures, 3 tables. Submitted to the IEEE International Conference on Communications (ICC) 2027

Hybrid precoding reduces the radio-frequency (RF) hardware burden in wideband massive MIMO-OFDM systems, but conventional optimization and deep neural network (DNN)-based approaches can incur high online complexity. This paper proposes XGB-SVD, a lightweight hybrid precoding framework that maps compact ray-tracing path parameters directly to analog beamforming codeword indices using cascade eXtreme Gradient Boosting (XGBoost). The cascade captures sequential beam-selection dependencies and transmit--receive propagation correlation, while the digital precoder and combiner are obtained from the reduced-dimensional equivalent channel using singular value decomposition (SVD). Evaluations over five DeepMIMO scenarios spanning sub-6-GHz and millimeter-wave bands show that XGB-SVD consistently outperforms the considered DNN baselines and approaches simultaneous orthogonal matching pursuit (SOMP) in several environments. Cross-scenario results further indicate improved transfer from propagation-diverse training environments, while XGB-SVD provides lower online computational and training costs than the DNN baselines.

0.37

Directional Bounded-Confidence Matrix Alignment in Vicsek Flocking

cs.RO ·2026-11-01 ·Hossein B. Jond
theoryunusual
Abstract

This paper proposes a directional bounded-confidence variant of the discrete-time Vicsek multiagent model, in which pairwise heading alignment is modulated by a state-dependent, positive semidefinite matrix that independently gates orthogonal radial and tangential components in the instantaneous line-of-sight frame. We analyze the normalized, matrix-weighted system dynamics, proving a uniform non-vanishing bound on proposal vectors that guarantees global well-posedness across arbitrary network states. For local small-spread states under complete metric interaction, we prove linear geometric contraction of the heading diameter and establish an offline-verifiable spatial bootstrap condition that guarantees metric contact preservation and coincidence avoidance a priori. A counterexample demonstrates invariant polarized states on connected matrix graphs, delineating systemic structural limits. Numerical experiments against unfiltered Vicsek averaging, scalar tied confidence, trace-matched scalar baselines, and restricted-angle rules demonstrate component-selective filtering without rejected-channel leakage, along with improved fixed-bound alignment on multi-seed benchmarks across varying population densities.

0.37

ODE Policy: Robot Learning via Action Dynamics, Not Actions

cs.RO ·2026-11-01 ·Jindou Jia, Jingliang Li, Kuangji Zuo, Tuo An, Yuxuan Hu +7 more
Abstract · 35 pages, 23 figures

Most robot policies, whether trained by imitation learning (IL) or reinforcement learning (RL), map observations to action values at discrete control steps and leave the continuity of physical motion to be learned implicitly. We propose ODE policy, which learns the dynamics of the action rather than the action itself. A neural ordinary differential equation (ODE) learns the vector field of the action by matching the integrated actions to the targets over short action chunks in IL and single control steps in RL. A lightweight diversity head captures multi-modal behavior by randomizing the direction of the field. At deployment, each action is integrated from the previous one, so the executed trajectory is continuous by construction. We evaluate it as a visuomotor policy, as the action head of vision-language-action (VLA) and world action models (WAMs), and as the actor of model-based RL against 14 baselines on unimanual, bimanual, dexterous, locomotion, and loco-manipulation tasks. In IL, against flow matching, it reaches 5.9$\times$ the success, 5.4$\times$ the generalization, and 3.9$\times$ the mode balance after one epoch, and generates actions 10.8$\times$ faster. It mostly moves more smoothly than human demonstrations. On 14 real-world tasks across single-arm, dual-arm, dexterous, and humanoid robots, it achieves the highest success rate. In RL, it further speeds up SuperMimic for humanoid motion tracking and smooths its actions. Efficient in learning, data, parameters, and execution, ODE policy offers a promising path to scaling robot learning.

0.36

Democratizing Robot-Use Agents: A Unified Harness with Open Models

cs.RO ·2026-11-01 ·Jiarui Yuan, Yuxin Zuo, Xiaoyu Huang, Wenbo Zhang, Kwanyoung Park +23 more
community pick
Abstract

Robot-use Agents (RUAs) have shown promising zero-shot manipulation capabilities, but reliance on frontier, proprietary models limits research into further training. To democratize RUA research, we introduce uni-robot-agent, a unified harness for competitive open-weight robot use that supports further training. Without privileged inputs, an out-of-the-box Qwen3.8-Flash-Next achieves a RoboDojo Score of 31.39, surpassing the published GPT-6 Astra-RoboProbe result of 28.97 and many VLA/WAM baselines, and performs competitively on 19 real-world tasks. Our analysis across model generations, families, and sizes links effective robot use to four general capabilities, suggesting that future RUA training should develop them jointly.

0.36

Directional Bounded-Confidence Cucker-Smale Flocking

cs.RO ·2026-11-01 ·Hossein B. Jond
Abstract

This paper introduces a directional bounded-confidence extension of the Cucker--Smale (CS) flocking model that decomposes pairwise relative velocities into independent radial and tangential channels within the instantaneous line-of-sight frame. Gating these components via compactly supported kernels creates a time-varying, state-dependent, matrix-weighted interaction that strictly generalizes classical scalar CS alignment. We establish an exact energy-dissipation identity with conserved mean velocity and prove exponential velocity consensus under a sliding-window persistent excitation (PE) condition formulated directly on the disagreement subspace. Furthermore, we derive a CS bounded-diameter criterion separating metric decay from aggregate directional coercivity, along with a cluster convergence guarantee. Simulations validate the exact dissipation balance and demonstrate that directional gating successfully restores global alignment in scenarios where traditional isotropic tied-confidence laws remain polarized.

0.36

History-Aware Force Sensing and Tunable Stiffness in a Pneumatic Kresling Origami Cell

cs.RO ·2026-11-01 ·Hengyi Zhou, Jiaqi Wang, Yilin Ma, Dezhong Tong, Xiaonan Huang
Abstract

Mechanical hysteresis and pneumatic loading complicate contact force estimation from deformation measurements in soft robots. This work presents a 3D-printed polyether block amide (PEBA) Kresling cell with magnetic angular sensing that supports three operating modes: force sensing at zero gauge pressure, force sensing with tunable stiffness under gauge positive pressure, and self-sensing actuation under negative gauge pressure. A multiscale gated recurrent unit (GRU) network estimates contact force from angular histories to account for the the combined effects of geometric deformation and material response. At zero gauge pressure, the model achieves a test RMSE of 0.76~N. Positive gauge pressure approximately doubles the average axial loading stiffness at +20~kPa over the first 3~mm of compression, while pressure-specific models maintain RMSEs below 0.93~N. Under negative gauge pressure, an energy-based compensation model enables contact force estimation using the the model trained at zero gauge pressure, achieving RMSEs of 1.66--2.10~N under quasi-static conditions. These results demonstrate a compact approach to combining force feedback, tunable axial stiffness, and self-sensing actuation in soft origami systems.

0.36

MagPillar: A Soft Magnetopillar Array Tactile Sensor for Robotic Surface Perception

cs.RO ·2026-11-01 ·Haotian Guo, Yu Feng, Hao Wu, Jiazhi Tang, Ningbin Zhang +1 more
Abstract

Robotic surface exploration benefits from tactile interfaces that accommodate contact and capture variations in surface characteristics. Herein, we report MagPillar, a tactile sensor inspired by spider tactile hairs that combines a magnetic pillar array with three-axis Hall sensing. A full-array mechanical and magnetic-field model characterizes how pillar deformation affects the superposed fields measured across the array, providing a physical basis for interpreting its contact responses. The compliant pillar architecture translates contact with surface textures and geometric features into distinct magnetic response patterns. Then, in robotic surface perception experiments, spectral features and logistic regression distinguish six material surfaces with an offline accuracy of 99.34% under within-session temporally blocked cross-validation. Continuous scans over periodic protrusions produce amplitude and peak-spacing variations consistent with changes in surface geometry. These findings connect compliant contact mechanics with distributed magnetic sensing, paving a new path for material recognition and qualitative surface-profile perception.

0.35

Charge-Balanced Control of Linear Systems: Controllability, Stabilizability, and Periodic Regulation

eess.SY ·2026-11-01 ·Jie Ruan, Zonglin Liu, Olaf Stursberg, Yuan Fan, Marcel van Gerven +1 more
unusual
Abstract · 14 pages

Electrical neuromodulation is a therapeutic approach that regulates neural activity by delivering electrical currents to brain tissue to treat brain disorders, such as epilepsy and Parkinson's disease. A critical safety requirement is charge balance, which ensures zero net injected charge over each stimulation cycle. Despite its clinical importance, the implications of charge-balanced stimulation for system dynamics and control remain poorly understood. This article investigates the controllability, stabilizability, and regulation of discrete-time linear systems under charge-balanced input constraints, motivated by applications in electrical neuromodulation. We introduce novel notions of charge-balanced controllability and stabilizability and establish necessary and sufficient conditions for both properties. We further demonstrate that nonzero steady states cannot be maintained under charge-balanced inputs, motivating the introduction of charge-balanced periodic orbits as alternative control targets. We characterize the conditions under which convergence to such orbits is achievable and develop optimal feedback controllers to stabilize the origin and track optimal charge-balanced periodic orbits. Finally, we validate the proposed control framework using a Hodgkin-Huxley neuron model, demonstrating successful suppression of tonic spiking and highlighting the potential of our approach for controlling nonlinear dynamical systems.

0.34

Beyond Ensemble Disagreement: Predictive STL Monitoring of RL Locomotion under Distribution Shift

cs.RO cs.LG ·2026-11-01 ·Qurban Ali, Badaruddin Chachar, Maqsood Ali, Sajjad Ali Rajper
Abstract · Accepted at the Modeling Uncertainty in Robotic World Models Workshop CoRL 2026

RL locomotion policies trained under nominal simulation conditions may encounter distribution shifts that compromise safety. Predictive monitors could anticipate violations, but their world-model forecasts may become unreliable under shift, and ensemble disagreement is not calibrated. We evaluate predictive Signal Temporal Logic (STL) monitoring for frozen Ant-v5 and HalfCheetah-v5 policies using closed-loop world-model rollouts and a three-model ensemble. Across external-force, observation-noise, and mass shifts, predicted robustness ranks HalfCheetah's future STL violations strongly, whereas disagreement does not reliably track observation prediction error or violations. Under severe noise, violations remain common ($64.2\pm24.4\%$ at $K{=}20$ across three PPO seeds), yet current-state robustness without prediction comes within $0.03$ AUROC of predictive robustness. Penalizing predicted robustness by disagreement does not consistently improve ranking. Ant provides complementary per-specification evidence, but its composite target has no positive violations and its any-violation target is single-class in most evaluated seed-condition slices. Overall, predictive robustness provides useful STL-specific ranking but little demonstrated additional ranking value over the current state; these findings caution against treating ensemble disagreement as a reliable indicator of safety uncertainty.

0.32

Metabolic heat and moisture reuse with climate-adaptive control for energy-efficient black soldier fly bioconversion in modular container farms

eess.SY ·2026-11-01 ·Sriram Saravanan, Philipp Sauerteig, Moritz Gold, Stefan Streif
unusual
Abstract · Submitted to Journal of Insects as Food and Feed. Submitted PDF consists of 6 figures

Black soldier fly larvae (BSFL, Hermetia illucens) convert organic residues into feed ingredients and fertiliser. Climate control is a major contributor to the energy demand of container-based bioconversion. We developed a control-oriented dynamic model linking larval growth, substrate dynamics and container climate to heating, ventilation and air conditioning (HVAC). A two-level Model Predictive Control (MPC) strategy planned climate trajectories and tracked them through HVAC actuation. Four simulation studies examined controller selection, HVAC configuration, larval age pairing and inter-container coupling, using subsets of eight climate scenarios. Across the four scenarios tested for controller selection, dynamic MPC reduced specific energy consumption (SEC) by 31 to 61% and increased predicted dry biomass at harvest by up to 27.5% relative to single fixed setpoints. Under the tested controller settings, active cooling was required in the two hot scenarios, whereas dehumidification provided no additional operational benefit. Pairing larval ages four days apart within one container reduced SEC by up to 33% in the tested non-hot scenarios. In the Cooldry scenario, two age-separated containers connected by an actuated air duct produced 9.0% more biomass and achieved 3.9% lower SEC than two independent containers with mixed ages, despite using 4.6% more total HVAC energy. The simulations suggest that dynamic climate control and metabolic heat and moisture reuse can improve the energy intensity of BSFL bioconversion. The quantitative estimates and hardware recommendations require experimental validation before application to industrial facilities.

0.31

Answering Without Arriving: An Evidence Audit for Embodied Question Answering

cs.RO cs.CV ·2026-11-01 ·Shubh Garg, Divyansh Pandey, Niyati Rawal, Vedant Khandelwal, Amit Sheth
benchmark
Abstract · 19 Pages, 3 figures, 11 tables

Embodied question answering asks an agent to navigate, observe, and then answer questions about what it saw, yet answer accuracy alone cannot tell whether any of this happened. A plausible answer can come from the question, the room label, or the answer choices, and a correct answer does not show that the supporting observation was ever made. We present a multi-room benchmark whose 200 test episodes and 800 questions are stratified by route extent, with an evaluation ladder that holds the questions fixed while varying only the evidence supplied to the answerer, from text-only probes through closed-loop agents to replay of the reference trajectory. Questions are generated from symbolic records of rooms, visible objects, and frames, so every answer can be traced to the observations that support it. A closed-loop vision-language agent reaches no target in 200 episodes yet answers 45.5% of questions, within 0.1 points of the same model given no images; replaying the reference trajectory raises the same model to 57.4%, so the evidence exists along the route but is not gathered. A family-level audit then shows that access to evidence is not the same as use of it. Four of eight families are largely answerable from text, one cites frames that cannot establish its own answer, and the oracle conditions reach 59.5% on Rooms-visited by counting the supplied images. Evidence acquisition, answer coverage, and shortcut sensitivity must therefore be established before an accuracy gain can be credited to memory.

0.31

Recursive Self-Improvement of Exoskeleton Assistance Policies through Human-in-the-Loop Feedback

cs.RO ·2026-11-01 ·Muyuan Ma, Shuo Jiang, Lijun Han, Xinpan Meng, Zhenghua Ma +3 more
hot direction
Abstract

To reduce reliance on manual design and extensive parameter tuning, a framework for the autonomous evolution of exoskeleton assistance policies through human feedback is proposed. Assistance policies are autonomously generated as executable control code by an intelligent agent and evaluated in experiments with human participants. Subjective user feedback, movement performance, and muscle activation measurements are incorporated into policy assessment. Based on these assessments, policy deficiencies are autonomously identified and the control code is revised by the agent for subsequent trials. A closed loop of policy generation, human feedback, evaluation, and autonomous refinement is thereby established. The framework is evaluated using GPT, DeepSeek, GLM, and Kimi, and initial assistance performance and refinement outcomes are compared across models. Despite differences in initial performance, effective assistance is achieved with all four models after a single autonomous refinement iteration under the tested conditions. Across three tasks, biceps and triceps muscle activation levels are reduced by 29.4% and 35.2%, respectively, compared with those measured without an exoskeleton. However, limited adaptability to unexpected situations is observed, indicating that further improvements in robustness and safety are required.

0.30

Distributed Pinching-Antenna Architecture for Communication and Target Positioning

eess.SP ·2026-11-01 ·Haoran Ni, Jiajun He, Mohammadali Mohammadi, Hien Quoc Ngo, Michail Matthaiou
publishedtheory
Abstract · Accepted for publication in the Proceedings of IEEE GLOBECOM 2026

We propose a pinching-antenna system (PASS)-assisted integrated sensing and communication (ISAC) framework for joint downlink communication and target positioning. The dielectric waveguides enable flexible signal radiation through activated pinching antennas (PAs), while the reflected echo signals are collected by a sensing receiver. A unified waveform is adopted to support both data transmission and sensing-echo processing. In particular, a two-stage procedure is developed for target localization using time-sum-of-arrival (TSoA) measurements, where the signal propagation distances are first estimated through de-chirp processing and beat frequency extraction. Then, the target position is obtained based on TSoA measurements using the weighted linear least squares(WLLS) technique. To characterize the fundamental limits of the PASS-assisted ISAC system, a Cramér-Rao lower bound (CRB) is derived, which captures the impact of PA activation. Furthermore, the PA activation problem is formulated to jointly optimize the communication spectral efficiency (SE) and sensing performance. To address this, a low-complexity heuristic greedy selection algorithm is proposed to determine the activated PA pattern. Numerical results demonstrate that the proposed scheme outperforms random PA selection, achieving gains of up to 81% improvement in communication SE.

0.30

FADE: A Flow-Based Approach for Multichannel Tail Estimation

eess.SP cs.AIeess.SY ·2026-11-01 ·Parmida Valiahdi, Niloofar Mehrnia, Mustafa Furkan Beker, Ubeydullah Erdemir, Ibrahim Hokelek +1 more
Abstract

Ultra-Reliable Low-Latency Communication (URLLC) is vulnerable to rare low-received-power events, and the spatial-diversity benefits of multiple receivers can be significantly reduced when they experience simultaneous deep fading. Multivariate Extreme Value Theory (MEVT) provides a principled framework for characterizing such joint tail behavior. However, conventional MEVT methods often rely on restrictive parametric assumptions or discrete representations of the dependence structure, which may not fully capture complex multivariate tail behavior. This paper proposes Flow-based Angular Density Estimation (FADE), a flow-based approach for characterizing the dependence between extreme events across receivers within the MEVT framework. FADE transforms multivariate extremes into a standard extreme-value representation and uses a flow-based density estimator to learn the dependence structure from the transformed data without imposing a predefined parametric form. The proposed framework is evaluated using experimentally collected data from a UAV communication system, in which the UAV serves as the transmitter and communicates with two ground receivers. FADE provides closer agreement with the empirical joint CDF, achieving reductions of up to 22.4% in MAE compared with the state-of-the-art Logistic and Maximum Empirical Likelihood (MEL) methods, while also improving distributional agreement and reducing absolute bias.

0.30

Multirate Data-Driven Predictive Control: Rank Obstructions, Identifiability, and Noise Conditioning

eess.SY math.OC ·2026-11-01 ·Robert Vrabel
Abstract

We develop rank, identifiability, and noise-conditioning results for compact input--output predictors in multirate data-driven predictive control. The inverse of a square zero-input response matrix used in an existing compact-lifting construction is shown to be undefined under its observability and minimal-polynomial assumptions. We derive exact replacements using rank-complete data, characterize all predictor extensions, and establish the minimum-norm and minimax Euclidean noise properties of the canonical Moore--Penrose extension. A joint predictor is identified directly from mixed experiments with arbitrary initial states. Deterministic perturbation bounds quantify reconstruction errors through informative singular-value margins, while a positive-definite reduced weighting preserves the nominal quadratic cost. Sampling-rate observability and event-triggering analyses identify additional conditions needed for reconstruction and stability guarantees.

0.30

Domain-Aware Hierarchical Federated Learning for UAV Swarms Across Heterogeneous Environments

eess.SP cs.LG ·2026-11-01 ·Taha Mahmood, Mudassar Shakeel, Ahmed Naeem, El Mehdi Amhoud, Rizwan Ahmad +1 more
Abstract

Hierarchical federated learning (HFL) enables communication-efficient training in unmanned aerial vehicle (UAV) swarms, but performance degrades under changing and heterogeneous environments. This paper proposes domain-aware HFL (DA-HFL), integrating mobility-informed clustering, cluster-level aggregation, cloud-based Fisher-weighted merging, a domain-adversarial shared backbone, environment-specific heads, and gated inference with local adaptation. A mission-weighted optimization formulation incorporates energy, latency, and cluster-head eligibility constraints. Theoretical analysis establishes resource feasibility, a conditional backbone-stationarity bound, the quadratic optimality of Fisher-weighted merging, and communication and computation complexity bounds. Simulations across four heterogeneous environments demonstrate faster convergence, improved accuracy during environmental transitions, and more stable, purer clusters than the evaluated baselines, with modest energy overhead over resource-aware HFL. Ablation studies validate the contributions of the proposed components.

0.29

A Data-Driven, Gain-Scheduled Equivalent Circuit for the B-H Behavior of Power Ferrites

eess.SY ·2026-11-01 ·Wilhelm Kirchgässner
Abstract

An equivalent circuit for the dynamic B-H behavior of power ferrites is identified from measured transients. It consists of classical elements of hysteresis modeling: a nonlinear inductor for the reversible curve, six viscous play branches (inductors in series with diode-bridge sliders) that form a Prandtl-Ishlinskii model, two R-L branches for the magnetic after-effect, a mean-field feedback of H as in the Jiles-Atherton model, and a measurement path. All elements are positive apart from the feedback sources, and the circuit can run in SPICE. Discretized at the sampling step, the circuit is a difference equation with 32 parameters per material, fitted by backpropagation through time. On the 15 materials of the MagNet Challenge 2, the mean sequence relative error is 7.33 % against 7.77 % for the challenge-winning recurrent network with 336 parameters, and the circuit is more accurate on 12 of the 15 materials. The accuracy can be scaled up by gain scheduling, where a small network per material sets the element values at every step from the excitation, while the hysteresis memory stays in the circuit, such that the error drops to 5.25 % with 656 parameters per material. Thus, an interpretable model is demonstrated that achieves state-of-the-art accuracy.

0.29

Adaptive Surface Cleaning under Uncertain Cleaning Response

cs.RO ·2026-11-01 ·Ruichen Li, Rachel Holladay
Abstract · 7 pages, 2 figures

Surface cleaning requires robots to adapt both where they clean and how they interact with contamination whose response to cleaning may vary spatially and remain unobservable prior to interaction. We formulate adaptive surface cleaning as a partially observable Markov decision process and propose a belief-aware planning framework that uses observed cleaning outcomes to update a latent local cleaning-response model and adapt subsequent cleaning motions, forces, and passes. Preliminary grid-world simulations demonstrate the resulting closed-loop planner adapting its spatial decomposition and cleaning actions as the surface evolves.

0.29

Low-Latency Joint Source-Channel Coding with Multiple Constant Curvature Curves

eess.SP ·2026-11-01 ·Anders M. Buvarp, Stefan Werner
Abstract

We propose constant-curvature curve multicode (C3M), a multi-branch extension of constant-curvature curve tube (C3T) analog joint source-channel coding for one-shot transmission over fixed isotropic AWGN channels. Instead of mapping a scalar source to a single curve on the power sphere, C3M partitions the source interval across a union of constant-speed branch curves in even-dimensional Euclidean space. This construction increases the available geometric degrees of freedom while preserving a low block length analog encoding structure. We derive the induced mixture output distribution, posterior representation, MAP and MMSE estimators, and mutual-information decomposition into branch-identification and along-curve components. We then introduce geometry- and classification-aware branch allocation rules based on tube density, curve length, and branch-classification reliability. The resulting design procedure fixes the optimal C3T curve as a starting branch and sequentially selects additional branches, thereby reducing the combinatorial cost of exhaustive multi-curve search. We use NORM-inspired branch-conditioned spectral decoders based on Fourier-feature lifting of the received ambient vector. Numerical experiments in four and six dimensions show that adding candidate branch curves improves the finite-dimensional end-to-end signal-to-distortion ratio, with selected two- and three-candidate C3M systems outperforming the one-curve baseline at +10 dB SNR. The results show that multi-curve analog mappings provide a practical finite-dimensional design route for low block length joint source-channel coding.

0.29

AI-Aided and AI-Oriented Integrated Sensing, Communication, and Computation: A Survey

eess.SP cs.AI ·2026-11-01 ·Yaya Etiabi, Wessam Ajib, El Mehdi Amhoud, Halima Elbiaze
survey
Abstract

The evolution from 5G to 6G networks marks a transition toward the "intelligence of everything." While 5G enables enhanced connectivity, 6G aspires to embed intelligence directly into communication infrastructures, creating systems that are adaptive, context aware, and capable of autonomous decision making. Realizing such systems depends on harmonizing three fundamental functions that compete for shared network resources and often pursue conflicting goals: (i) sensing for environmental and contextual information acquisition, (ii) communication for reliable and timely information exchange, and (iii) computation for information processing and decision making. This harmonization necessitates shifting from independent optimization to their joint design as a unified system, a paradigm known as Integrated Sensing, Communication, and Computation (ISCC). In parallel, AI has emerged both as an enabling tool for ISCC optimization and as a workload whose requirements influence ISCC system design. This survey comprehensively examines ISCC from this dual perspective by distinguishing between AI-aided ISCC, in which learning-based methods support the joint design and optimization of sensing, communication, and computation, and AI-oriented ISCC, in which these functions are jointly configured to support learning, inference, and distributed intelligence. The survey further examines the co-design of AI and ISCC, where AI requirements and ISCC constraints jointly shape system and model decisions. It further provides a comparative synthesis of the resulting design approaches, system architectures, performance metrics, and evaluation practices, relates these developments to application-specific requirements across different domains, and identifies open research problems toward scalable, adaptive, and trustworthy AI-ISCC systems.

0.28

Green Learning-Based Signal-Level Detection of RIS Spoofing Attacks in Vehicular ISAC Systems

eess.SP cs.IT ·2026-11-01 ·Po-Heng Chou, Tingyu Shui, Pin-Qi Fu, Lien-Sheng Chang, Walid Saad +1 more
Abstract · 6 pages, 4 figures, and 2 tables. Submitted to the IEEE International Conference on Communications (ICC) 2027

Integrated sensing and communication (ISAC) enables wireless infrastructure to jointly support communication and environmental sensing, but also introduces new physical-layer security threats. A malicious reconfigurable intelligent surface (RIS) can manipulate its time-varying reflection phases to generate spoofing echoes and distort the sensing results of a roadside unit (RSU). Existing trajectory-level detectors rely on reconstructed sensing parameters and vehicle trajectories, making their performance dependent on upstream estimation accuracy. In this paper, we propose a signal-level Green Learning (GL) framework that directly detects RIS spoofing from matched-filtered RSU echoes. The proposed method combines two-stage Subspace Approximation with Adjusted Bias (SAAB), Discriminant Feature Test (DFT), and eXtreme Gradient Boosting (XGBoost) for spatial-temporal representation learning, feature selection, and classification, respectively. Simulation results show 99.335\% detection accuracy for a 32-element malicious RIS at $-80$ dBm receiver noise power. GL remains competitive with Long Short-Term Memory (LSTM) and bidirectional LSTM (BiLSTM) baselines while requiring only 0.1119 million floating-point operations (FLOPs), and also generalizes to previously unseen RIS sizes.

0.28

A Global Routing Method for Electronic Circuit Design Integrating Flow Matching and Constrained A*

eess.SY ·2026-11-01 ·Sanghoon Jin, Sumin Lee, Namwoo Kang
Abstract

Global routing is a fundamental issue in the design of electronic circuits, as it directly relates to the efficiency, reliability, and compactness of the overall layout of integrated systems. As the demand for higher functionality and reliability within a limited chip area increases, the importance of routing as a design consideration has never been greater. Efficient global routing optimization techniques are therefore required to connect all pins with minimal wirelength (WL), while avoiding congestion and interference across multiple nets within the design space. As the number of nets to be routed increases, global routing inherently suffers from a combinatorial explosion in the design space, making it increasingly difficult to obtain feasible solutions. In this paper, we focus on solving constrained global routing problems based on a heuristic A*-based framework. Based on this concept, a hybrid approach is proposed by integrating constrained A* with flow matching (FM). The method utilizes FM to construct a global cost map that captures routing tendencies across the design space, which is then used to guide the constrained A* search process. This enables efficient exploration of the solution space while satisfying routing constraints. Experimental results demonstrate that the proposed method achieves comparable WL and via usage to baseline methods while maintaining constraint satisfaction across a wide range of scenarios. Notably, the framework shows strong capability in solving complex routing instances that are difficult to handle using conventional approaches. Additionally, the method shows robust generalization performance across different problem sizes and net configurations.

0.27

When to Probe and When to Act: Clarity-Aware Risk-Adaptive Control for Interactive Driving

cs.RO ·2026-11-01 ·Youran Wang, Devodita Chakravarty, Yiwei Lyu
position
Abstract · 8 pages, 4 figures, 2 tables. Submitted to ICRA 2027. Youran Wang and Devodita Chakravarty contributed equally

Safe navigation in interactive traffic requires autonomous vehicles to reason about unobservable human characteristics and the reciprocal nature of multi-agent interactions. Traditional architectures often decouple prediction from planning, leading to the "Frozen Robot Problem" or unsafe maneuvers under high ambiguity because they treat risk assessment and information acquisition as independent objectives. We argue that the notion of safety should not remain static but must instead dynamically co-evolve with the system's internal confidence: the robot should actively clarify ambiguity when necessary while also recognizing the diminishing returns of information, and naturally transition toward task execution. We introduce a unified belief-space Model Predictive Control (MPC) framework that couples multistep information planning with a clarity-adaptive Conditional Value-at-Risk (CVaR) barrier mechanism. By treating clarity, a scalar informational state, as an explicit planning variable, the controller co-designs its probing strategy with its safety specification, tightening conservatism under high ambiguity and relaxing constraints as confidence improves. We further provide a performance analysis showing that the cost of information seeking and the expected planning-cost gap to a controller that knows the true human characteristic are bounded by residual ambiguity and solver error. We evaluate the framework in lane-merging and unsignalized-intersection scenarios against representative interactive-planning baselines. Across 5,000 simulations, the proposed method achieves the highest success rates (94.6% and 93.2%) and the lowest collision rates (1.2% and 1.4%) among the compared methods, providing a principled foundation for robust decision-making in human-centric environments.

0.27

Control Barrier Functions for Safety-Aware Latent-Space Motion Planning

cs.RO cs.AI ·2026-11-01 ·Chandrakiran Ravula, Siddhanth Ramesh
Abstract · 8 pages, 2 figures

Latent-space path planning (LSPP) compresses the configuration space of a robotic manipulator into a compact, differentiable manifold learned by a variational autoencoder, enabling fast gradient-based planning without explicit inverse kinematics. Existing LSPP frameworks enforce obstacle avoidance with a binary collision classifier, whose saturated gradients near obstacle boundaries and lack of temporal structure cause the optimizer to stall and provide no forward-invariance guarantee. We replace the classifier with a neural control barrier function (CBF) learned directly in the latent space. The barrier is trained with a three-term objective that combines safe and unsafe sign constraints with a discrete-time decrease condition evaluated on transitions of an obstacle-blind nominal planner, and it is enforced at planning time by a closed-form, first-order correction that projects each goal-directed latent step back onto the CBF constraint. On a 7-DoF Franka Panda manipulator, the CBF planner achieves a 75.6\% success rate on a fixed-obstacle task, 12.0 percentage points above the best of 35 tuned classifier configurations, and 83.8\% on a multi-position task. A sweep of the decay rate $α$ characterizes the safety--performance trade-off, with the best operating points in $α\in [0.10, 0.20]$. In a fully general random-obstacle setting, success drops to 33.0\%, which we trace to covariate shift between the obstacle-blind training transitions and the states visited by the corrected planner.

0.26

Clock Offsets and Ranges from Optical Pseudoranges

eess.SP ·2026-11-01 ·Christoph Günther
Abstract

This note describes the estimation of ranges and clock offsets from measured pseudoranges in the context of the Kepler Navigation System. The latter system continuously adapts the frequency of its clocks to ensure that they all show the same time independently of where they are. In Medium Earth Orbits, the joint problem of estimating ranges and clock offsets can then be factorized.

0.26

HuG: Learning Humanoid Whole-Body Affordance-Aware Grasping with Active Perception

cs.RO ·2026-11-01 ·Xuanbin Peng, Lai Wei, Yuze Zhai, Qingyuan Yu, Keyi Wang +2 more
code/data
Abstract · CoRL 2026. Project page: https://xuanbinpeng.com/hug

Acquiring objects in human environments is hard for humanoids: targets are partially occluded under egocentric viewpoints, and reaching them demands coordinated whole-body motion that keeps the target visible and reachable while the base stays stable. We present HuG, a framework for whole-body affordance-aware grasping with active perception that couples perception and action through a Perceive-Align-Grasp pipeline: belief-driven active perception resolves occlusion by treating viewpoint as an explicit decision variable; a point-cloud-conditioned affordance policy proposes whole-body grasps; and an affordance-aware alignment stage repositions the body to the most executable configuration. A key insight is that, on a floating base, kinematic validity is not enough: we curate training data with controller-in-the-loop dynamics simulation, rejecting 73% of kinematically valid grasps that would destabilize the deployment controller, so the policy learns kinodynamically feasible affordances. Trained entirely in simulation with zero real demonstrations, HuG attains 87.5% real-world success on a multi-level shelf task, generalizes to novel objects, and holds at least 73% success at every tested height. Beyond generalizing within its training distribution, the policy extrapolates spatially: at scale in simulation it succeeds at table heights well below any seen during training, indicating that kinodynamically filtered synthetic data grounds affordances in object geometry and whole-body feasibility rather than in memorized configurations.

0.26

Serving Agentic Robot Fleets: Workload Characterization, KV-Cache Thrashing, and Admission Control

cs.RO cs.DC ·2026-11-01 ·Sida He
Abstract · 12 pages, 17 figures, 8 tables

Agentic robots combine a vision-language brain, a tool-using harness, and a vision-language-action (VLA) policy. Serving a fleet of these agents introduces two coupled resource constraints: the brain repeatedly revisits long, growing key-value (KV) cache states, while the VLA executes action chunks under tight control deadlines. We characterize 80 LIBERO-PRO episodes generated by GPT-6-Astra, RPent, and pi0.5 (1,862 brain turns and 1,582 VLA inference chunks), profile both layers on H800 graphics processing units (GPUs), and replay the resulting agent programs on a vLLM serving engine. Brain contexts reach 215k tokens; the median Astra request takes 7.78 s, while the intervening environment-step gap is only 1.0 s. Under a 0.25 s action-chunk deadline, one VLA GPU meets the 1% starvation target for one robot; under a 1.0 s deadline, its measured capacity reaches at least 32. In sustained brain replay, increasing concurrency from 20 to 32 robots reduces throughput from about 2.4 to 0.6 requests/s as repetitive prefill displaces generation. A static active-episode cap restores 2.24-2.25 requests/s and raises completed trace-program throughput from 1.30 to 5.00-5.55 episodes/min. Our measurements expose the workload-dependent shift between action deadlines and brain KV residency, and establish practical principles for capacity planning and admission in sustained agentic robot fleets.

0.25

Human Sensorimotor Model Adaptation to Account for Robot Accuracy and Speed Augmentation in Human-Robot Co-Manipulation

cs.RO ·2026-11-01 ·Frits van Hall, Kevin Haninger, Luka Peternel
Abstract · 10 pages, 6 figures

In human-robot co-manipulation, the robot's trajectory should accommodate the human's, where human sensorimotor behavior is affected by the task's spatial and temporal constraints, as well as personal factors. Several robot trajectory planning methods have been proposed to embed human sensorimotor models in the robot trajectory planning, but these methods do not consider temporal constraints that adapt to how the human behaves when interacting with the robot. To better align human and robot collaborative motion, this work adapts a stochastic optimal control problem with a final state constraint from the task constraints and a temporal constraint from Fitts' law. A recursive least squares method is used to estimate the Fitts' law parameters, adapting the robot's motion duration to that of the human, thus allowing the robot to adapt to how the human behaves when interacting with the robot. It is hypothesized that a speed increase can result from the robot augmenting the accuracy. The feasibility of the approach is validated on a positioning task where the robot controls degrees of freedom which have repeatable motion, and the human adapts motion in the complementary degrees of freedom. The approach is shown to be computationally suited to online use, it converges as the human and robot mutually adapt, and can increase the movement speed, although thorough human studies are needed in future work.

0.23

$TF^2$-Steering: Action Refinement with Tactile Forecasts and Feedback via Velocity-Field Steering

cs.RO ·2026-11-01 ·Yiming Chen, Xiaofei Wei, Chun Gu, Wei Xiao, Peidong Liu +1 more
code/data
Abstract · 19 pages, 6 figures. Project page: https://yiming-cc.github.io/tf2-steering/

Tactile sensing plays a crucial role in everyday human interaction. Consider inserting a plug into a tight socket, vision alone provides limited guidance during the final insertion stage, and we instead rely on both anticipated contact and realized tactile feedback to ajust the insertion motion. The same capability is equally important for robots, particularly in dexterous manipulation tasks. Yet many contemporary robot policies remain vision-dominant. This visual bias, compounded by limited tactile data and sparse contact events, can cause directly incorporated tactile signals to be ignored or even compromise the policy's pretrained capabilities, yielding degrading performance as observed in our experiments. We present $TF^2$-Steering, which refines actions via learned residual steering of the flow-matching action expert's velocity field, using a tactile mixture-of-experts architecture to integrate both tactile forecasts and real-time tactile feedback. Together with specific designed three stages training recipe, $TF^2$-Steering enables tactile signals to correct and, when necessary, redirect the flow-matching action-generation process, while building on the strong action-generation capacity of the vision-based foundation policy. Extensive experiments across six simulated and six real-world manipulation tasks demonstrate the effectiveness of $TF^2$-Steering. It outperforms the strongest evaluated baselines by 10.7 percentage points in average simulation success rate and 9.1 percentage points in average real-world progress success rate. Project page: https://yiming-cc.github.io/tf2-steering/.

0.23

CompliantBot: Learning Whole-Body Compliance across Task Interfaces for Humanoid Control

cs.RO ·2026-11-01 ·Haoyang Li, Keqi Zhu, Zhanxiang Cao, Yang Zhang, Yizhi Chen +2 more
Abstract · 12 pages, 8 figures, 5 tables

Humanoid controllers need the ability to yield to external forces while retaining the capacity to resume their original tasks. However, tracking objectives penalize the deviations needed for yielding, while relaxing tracking goals persistently can make task recovery harder. Existing approaches often rely on task-specific interaction command interfaces or constructed compliant motion data and reference trajectories, limiting the direct reuse of their compliance-training mechanisms across different task interfaces. We introduce CompliantBot, a compliance-training framework for motion tracking and velocity-command locomotion. It preserves the original task inputs and requires no preconstructed compliant data or target trajectories. A state-dependent interaction force field couples applied loads to the robot's motion, enabling local articulation and whole-body movement to change the forces experienced. A complementary force-magnitude reward encourages lower interaction loads, while task rewards are downweighted during loading and restored after release to coordinate yielding and task resumption. The trained policies use task inputs and proprioceptive history without explicit external-force measurements. Across both task interfaces, CompliantBot yields farther under constant pulls and experiences lower spring interaction loads than matched baselines, while resuming task tracking after force removal. CompliantBot achieves the highest task completion among the compared methods under stronger pulls. Power evaluations further show that CompliantBot has the lowest mean joint mechanical power among the compared methods. Real-robot demonstrations further illustrate compliant responses through arm motion, stepping, and body reorientation.

0.23

Dual Path Following for an Aerial Manipulator Under State and Input Constraints via Transverse Feedback Linearization

cs.RO eess.SYmath.OC ·2026-11-01 ·Hamza Tariq, Adeel Akhtar
code/data
Abstract · 8 pages, 3 Figures

This paper studies dual path following for a quadrotor carrying a three-degree-of-freedom arm. The center of mass of the combined system must follow a vehicle path and the end effector must follow a second path, while the arm respects its joint limits and avoids its singular postures. The vehicle and the arm are mechanically coupled, so one controller must handle both paths. With the center of mass and the end-effector path coordinates as outputs, and a dynamic extension of the thrust, the system admits a normal form without internal dynamics. We characterize the states where this normal form fails. We formulate the transverse feedback linearizing controller as a quadratic program that relaxes only the speed of the end effector along its path and imposes safety and actuator limits as inequality constraints. The program reduces to a scalar problem with a closed-form solution. If it remains strictly feasible for all time, the feedback is locally Lipschitz, both paths are invariant and attractive, and every constraint holds. In the physics engine Drake, the controller resolves a conflict between the two speed references by slowing the end effector while both paths remain invariant. By contrast, a cascade with a safety filter leaves both paths. Code and animations are available at https://gradslab.github.io/am_dual_pf-/

0.22

Do Embodied VLMs Actually Use Their Scene-Graph Memory?

cs.RO cs.AIcs.CV ·2026-11-01 ·Shubh Garg, Divyansh Pandey, Vedant Khandelwal, Niyati Rawal, Amit Sheth
benchmark
Abstract

Embodied agents that answer questions about their route need a memory of what they saw, and scene graphs are a natural form for it. A vision-language model (VLM) can receive such a graph as text or through a graph encoder, but accuracy alone cannot show whether the graph holds the episode's facts, reaches the model, or is actually relied on. In this diagnostic study, we introduce a graph-swap test for the last of these: each episode is given another episode's graph, and we count how many answers change. On an embodied question-answering benchmark with routes through two to seven rooms, we use it to compare ways of delivering scene-graph memory to a fine-tuned VLM. As text, the memory is used strongly, adding 30 points on route questions. As a graph, it is not, and we trace why: the encoder preserves the information, but standard pooled and late fusion discard it. Giving each node its own input token recovers part of it, a gain of 8.3 points on route questions that the swap test confirms comes from the graph. The test also exposes a shortcut that accuracy hides: when the same facts appear in the text, the model ignores its graph. The test thus offers a cheap check of whether a VLM uses its scene-graph memory, and our analysis points to how that memory should be delivered.

0.22

Spectral-Convolutional Factorized Transformer for Joint Multipath Parameter Estimation: Applications to mmWave Positioning and Radio SLAM

eess.SP cs.AI ·2026-11-01 ·Yaya Etiabi, Wessam Ajib, Halima Elbiaze
Abstract

Future 6G wireless networks are expected to evolve toward AI-native systems that support intelligent, environment-aware services, including positioning and mapping. Current research has already integrated AI into several aspects of communication, particularly wireless channel estimation. However, most learning-based channel estimators focus on communication-oriented objectives rather than explicitly estimating multipath parameters such as delay, angle of arrival (AoA), and angle of departure (AoD). These parameters provide essential geometric information for downstream tasks including positioning and radio simultaneous localization and mapping (SLAM). In this work, we propose a Spectral-Convolutional Factorized Transformer (SCFT) for joint multipath component (MPC) estimation from noisy complex channel estimates and physical beam-pair descriptors. SCFT combines local spectral convolution, permutation-equivariant beam-set attention, and a query-based Transformer decoder to detect and classify line-of-sight and non-line-of-sight paths and estimate their observed delays, AoAs, and AoDs. Experiments on realistic ray-tracing data show that SCFT reduces the generalized optimal sub-pattern assignment error by 13.1% compared with ResNet, the strongest competing baseline. We further demonstrate this advantage in independent snapshot positioning and radio SLAM, where SCFT reduces position RMSE by 18.5% and 65.0%, respectively, relative to ResNet. These findings highlight the importance of explicitly recovering multipath geometry, as improvements in MPC estimation can translate into substantial gains in downstream tasks.

0.22

Label-Free Performance Estimation of Deep Learning Based Radar Detectors in Time-Varying Environments

eess.SP ·2026-11-01 ·Chuanfei Zang, Yiru Lin, Yumiao Wang, Xingyu Chen, Xiaobo Yang +1 more
Abstract

Deep-learning-based radar target detectors can achieve strong performance in complex sensing environments. However, in time-varying environments, changes in clutter characteristics may cause the distribution of echoes to deviate from that of the training data, thereby degrading detector reliability. Timely identification of such degradation requires an accurate assessment of the deployed detector's performance; however, ground-truth labels for radar echoes are typically unavailable during deployment. This paper investigates label-free performance estimation for deployed deep learning-based radar target detectors. We first show that confidence-based estimators tend to overestimate detection performance under distribution shift because they fail to account for the discrepancy between the detector-induced pseudo-label distribution and the true label distribution. To address this limitation, we propose probabilistic optimal transport (POT), which assesses deployment performance by computing the Wasserstein distance between the detector's output probability distribution on unlabeled deployment echoes and the label distribution in the training data. We further develop thresholded probabilistic optimal transport (TPOT), which suppresses the influence of anomalous transport costs on the Wasserstein-distance-based estimate, thereby improving estimation robustness. Experiments on a publicly measured radar dataset demonstrate that POT and TPOT consistently outperform representative label-free baselines across deployment scenarios with varying degrees of data distribution shift.

0.22

OpenNWM: A Generalist Navigation World Model with Latent Action Pretraining

cs.RO cs.AI ·2026-11-01 ·Dujun Nie, Ruiqi Song, Zixuan Wang, Hanqing Wang, Yifan Han +4 more
code/data
Abstract · Project page: https://openspace-lab.github.io/OpenNWM/ Code: https://github.com/OpenSpace-Lab/OpenNWM

Navigation world models predict future observations conditioned on navigation actions to support planning, but their generalization across real-world environments remains constrained by the limited coverage of action-labeled data. Action-free navigation videos provide diverse visual experience, yet lack the recorded actions required for action-conditioned training. We introduce OpenNWM, a generalist navigation world model with latent action pretraining. It learns universal latent actions from large-scale action-free videos through visual prediction pretraining, and further aligns them with physical motions via action-conditioned post-training. To support video pretraining, we curate NavAnywhere, a large-scale action-free navigation video dataset comprising 17.5 million visual observations from 15 heterogeneous sources. It captures navigation by ground robots, humans, and drones, with diverse viewpoints and motion patterns across five broad scene types: residential areas, public and commercial spaces, urban settings, parks, and natural landscapes. Experiments demonstrate improved visual prediction on in-domain and out-of-domain benchmarks, together with improved navigation performance. These results show that latent-action pretraining enables diverse, action-free videos to improve dynamics learning and generalization for open-world navigation.

0.21

Differentially Private Distributed Optimization via the Intrinsic Randomness of Zeroth-Order Estimators: Sampling Geometry, Convergence, and Privacy Guarantees

eess.SY math.OC ·2026-11-01 ·Jiajun Wang, Yujie Tang
Abstract

Distributed zeroth-order optimization has gained increasing attention in applications where exact gradients are computationally expensive or unavailable. However, the iterative exchange of local decision variables makes distributed networks vulnerable to inference attacks. While differential privacy provides a rigorous framework for privacy protection, conventional approaches generally require additional random noise, which inevitably affects optimization accuracy. In this paper, we investigate whether the randomness inherent in zeroth-order estimation can itself provide privacy protection. We develop a distributed zeroth-order gradient descent method that integrates consensus updates with a ball-uniform two-point gradient estimator. Our theoretical analysis shows that our algorithm attains standard convergence guarantees in both online and offline optimization settings. We then characterize the relevant probability distributions of the associated privacy mechanism, and establish differential privacy guarantees for both the online and offline optimization settings. These results reveal the potential of intrinsic zeroth-order randomness for privacy protection.

0.20

Finite Global $H_{\infty}$ Optimization of the $Q$-Filter Time Constant for Nominal-Response Recovery in SISO Disturbance Observers under Parametric Plant Uncertainty

eess.SY ·2026-11-01 ·Hyuntae Kim, Wen-Hua Chen
theory
Abstract · 14 pages, 2 figures, 1 table

This paper develops a finite real-algebraic procedure for global H-infinity optimization of the time constant of a fixed-shape Q-filter in a single-input/single-output (SISO) disturbance observer (DOB) under affine parametric plant uncertainty. Robust internal stability defines a possibly disconnected feasible domain, and the objective is the worst-case weighted mismatch between the uncertain DOB reference response and the nominal closed-loop response. A recovery-error factorization preserves affine dependence of the performance numerator and denominator on the same uncertainty vector. Phase lifting and a complex-polynomial Edge theorem reduce the uncertainty box to finitely many polynomial edges, while quadratic dependence of the squared frequency-domain inequality on the edge coordinate permits analytic elimination of that coordinate. The resulting semialgebraic formulation has only the squared frequency as a universally quantified variable. It determines the global infimum, decides whether it is attained, and constructs the complete optimizer set over the full robust-stability domain. A DOB case study with a plant family containing non-minimum-phase members yields a unique optimal time constant of approximately 0.0364075 and a worst-case norm of approximately 2.29975. The optimizer lies strictly inside the first robust-stability component, and every smaller feasible time constant in that component has a larger worst-case norm. Finite algebraic certificates establish robust stability, performance feasibility, and strict global optimality.

0.20

GoalDex: Learning Image-Goal-Conditioned Policies for Dexterous In-Hand Manipulation

cs.RO ·2026-11-01 ·Arjun Bhardwaj, Mayank Mittal, Maximum Wilder-Smith, Vaishakh Patil, Marco Hutter
Abstract

Dexterous in-hand manipulation is commonly formulated around explicit object pose estimation, yet reliable pose tracking is challenging under severe hand-object occlusions and often requires object-specific supervision. This dependence creates a bottleneck between perception and control that can limit robustness and scalability. We present an end-to-end visuomotor framework for dexterous in-hand reorientation that operates on RGB observations and proprioception without intermediate pose estimation. Our approach jointly learns visual representations and control policies through behavior distillation from privileged state teachers and auxiliary reconstruction objectives. Furthermore, we introduce image-goal-conditioned reorientation, allowing desired object configurations to be specified directly through images rather than canonical pose representations. We evaluate our approach across a diverse set of object geometries and instances in simulation, where it consistently outperforms pose-based methods and policies built on pretrained visual encoders. Finally, we demonstrate direct sim-to-real transfer of a single image-goal-conditioned policy across five real-world objects, showing that accurate dexterous reorientation can be achieved without explicit pose estimation. For videos and code release, please refer to the project website: https://rffr.leggedrobotics.com/works/goaldex/

0.17

Mission Capability-Aware UAV Swarm Coordination for Autonomous Disaster Response

eess.SP ·2026-11-01 ·Mudassar Shakeel, Taha Mahmood, Ahmed Naeem, Muhammad Hamza Jehangir, El Mehdi Amhoud +1 more
Abstract

This paper proposes an artificial intelligence (AI) driven mission capability aware unmanned aerial vehicle (UAV) swarm coordination framework for autonomous post disaster search and rescue. An integrated sensing and communication enabled master high altitude platform station senses the disaster environment, detects victims and hazards, and constructs adaptive rescue regions based on spatially varying mission demands. A mission capability requirement (MCR) quantifies regional capability demand, while a normalized mission capability score (MCS) characterizes the real time capability of individual UAVs. A capability-aware assignment mechanism matches UAV capabilities to regional requirements while minimizing deployment distance. An MCS based reallocation mechanism dynamically transfers UAVs from capability surplus to capability deficient regions while preserving donor sufficiency. A hierarchical deep reinforcement learning framework with proximal policy optimization performs high-level region selection, while MCR MCS matching enables feasible UAV level reallocation. Finally, joint trajectory optimization considers travel time, distance, propulsion energy, operational capability, obstacle avoidance, and inter UAV collision avoidance. Theoretical analysis and simulations demonstrate improved mission capability satisfaction, resource efficiency, and disaster response performance

0.16

When Delay Meets Drift: Physical Safety of Robot Foundation Models Is a Cross-Layer Problem

cs.RO cs.CVcs.ET ·2026-11-01 ·Liangkai Liu, Bimal Thapa Magar
position
Abstract · 11 pages, 3 figures

Robot foundation models now act in a closed loop, yet their safety is evaluated one variation at a time: robustness benchmarks perturb the scene while assuming instantaneous inference, and real-time execution methods study inference delay in nominal scenes. We argue that physical safety is not a property of the model alone, but emerges from how variations in the physical environment, the model, the computing platform, and execution interact in closed loop. We formalize this as cross-layer variation propagation across these system layers and use an interaction contrast to measure when joint variation creates more risk than the sum of its parts. In 4,000 paired closed-loop episodes of the pi0.5 vision-language-action model on LIBERO, a 400 ms inference delay raises the safety-violation rate by 3.5 percentage points under the training viewpoint but by 37 percentage points under a 60-degree camera shift: 33.5 percentage points (95% CI: 25-43) more than the two effects combined. We close with a research agenda toward predictable physical intelligence: predicting when variation becomes risk, and adapting compute, power, and control accordingly.

0.14

Joint Subcarrier and Power Allocation for Multi-User OFDM ISAC Systems

eess.SP ·2026-11-01 ·Jiayue Huang, Ning Wei, Min Qiu, Aimin Tang
theory
Abstract

Integrated sensing and communication (ISAC) is envisioned as a key enabling technology for 6G networks. Efficient resource sharing between communication and sensing is a fundamental challenge in orthogonal frequency-division multiplexing (OFDM) ISAC systems, particularly in multi-user scenarios, where channel diversity can be exploited to improve resource utilization. This paper investigates joint subcarrier assignment and power allocation for a downlink multi-user OFDM ISAC system, with the objective of minimizing the delay Cramér-Rao lower bound (CRLB) subject to constraints on the normalized integrated sidelobe level (ISL) over the region of interest (RoI), per-user data rates, and total transmit power. We first derive a convex formulation for sensing power allocation given fixed subcarrier assignments, and then develop an equivalent mixed-integer conic formulation of the joint optimization problem. To obtain computationally tractable solutions, we propose a conic relaxation-guided fix-and-optimize (CRG-FO) algorithm that integrates continuous relaxation, power re-optimization under fixed subcarrier assignments, and restricted neighborhood search. Simulation results demonstrate that joint subcarrier and power allocation reduces the distance CRLB relative to uniform subcarrier and power allocation across the tested ISL thresholds, rate loads, and user configurations. Power optimization on uniform sensing supports captures most of this gain, while adaptive subcarrier assignment provides only modest additional improvement.

0.14

Local Solution Repair for Multi-Agent Pickup and Delivery

cs.RO cs.MA ·2026-11-01 ·Ethan Schneider, Jingkai Chen, Seth Hutchinson, Sonia Chernova
Abstract

Multi-robot systems in automated warehouses must efficiently attend to a constant stream of pickup-and-delivery tasks while avoiding collisions. Existing Multi-Agent Pickup-and-Delivery (MAPD) solutions generate plans that are globally feasible but often locally inefficient. In large-scale, long-horizon deployments, these inefficiencies, such as robots backtracking from target goal locations and idling in congested areas, accumulate over time and degrade performance. This paper contributes Local Solution Repair (LSR), a goal replanning approach that detects suboptimal behaviors at run-time and locally improves plans to reduce conflict between agents. We demonstrate the utility of our approach by repairing solutions generated by state-of-the-art MAPD baselines, M2M and LNS-PBS. Our results in long-horizon warehouse simulations over three environments show upwards of 5.9% improvement in task throughput over current state-of-the-art methods by reducing the number of suboptimal instances by ~36k or by 14.6% with minor additional computation time.

0.12

SCOUT: Planning under Occlusion via Object-Centric World Model Rollouts

cs.RO cs.AIcs.CV ·2026-11-01 ·Amir Hossain Raj, Xuesu Xiao
Abstract

Robots manipulating in cluttered spaces must often plan from partial views where foreground objects occlude the target, surrounding geometry, and usable free space. Reactive policies often miss these hidden constraints, while exhaustive search becomes computationally prohibitive in dense clutter. We present SCOUT, an occlusion-aware planning framework that bridges high-level deliberation with grounded physical rollouts. SCOUT leverages an object-centric scene representation to reconstruct hidden geometry and derive a relational planning state capturing blocker hierarchies and accessibility corridors. To navigate the vast action space in clutter, an LLM proposes structured action candidates which are then rigorously evaluated via Monte Carlo Tree Search (MCTS) using a learned action-conditioned world model. Evaluations on 'Reveal' and 'Placement' tasks show that SCOUT outperforms Vision-Language-Action (VLA) policies, embodied agents, and 3D world models, with significant gains in long-horizon tasks requiring reasoning over occluded geometry. We further validate the framework's robustness through real-world robotic manipulation in a cluttered shelf environment.

0.09

Same Path, Less Time: Training-Free Fast Execution of Vision-Language-Action Models under Inference Latency

cs.RO cs.LG ·2026-11-01 ·Fuyi Chen
Abstract · 22 pages, 11 figures

Vision-language-action (VLA) models predict chunks of future actions, yet deployed robots waste time in two ways that are independent of model capability: they halt while the next chunk is inferred, and they move at the speed of slow, cautious demonstrations. Existing remedies remove one loss each, and stacking them fails, because asynchrony and speed-up each change the correspondence between planned progress and executed motion. We present Path-preserving Time Scaling (PTS), a training-free execution layer for frozen action-chunking policies built on one principle: the robot's motion is a time reparameterization of a reference path, and the path changes only in response to new observations. A path layer decodes every replan deterministically in one step and enters it at the progress consumed during inference. A time layer sums the planned increments, keeps the arm at 1x while the gripper moves, and caps the rate by the actuator's linear range and by chunk feasibility under latency. We show that misaligned entry and stochastic sampling add a bias and a variance to the continuation error, and derive a break-even success rate that states when speeding up pays off. On LIBERO with two frozen VLAs under injected inference latency, PTS raises task throughput over the strongest deployable baseline by 33% for pi0.5 and 104% for SmolVLA while matching or improving the success rate, is best on all six model-suite combinations, and exceeds zero-latency 1x execution. Against asynchronous 1x execution with the same one-step decoder, the time layer alone adds 28% and 38%. On LIBERO-Dynamic it improves on stop-and-wait execution by 38% and is comparable to asynchronous 1x execution.

0.08

People & Society 22

The Cryptocurrency Trend Factor Revisited: Predictive Information, Portfolio Construction, and Pricing Claims

cs.CE ·2026-11-01 ·Yifei Wang, Zhenling Pang
published
Abstract

We reassess five economic claims about the cryptocurrency trend factor (CTREND): cross-sectional return predictability, high risk-adjusted performance, incremental information beyond market, size, and momentum, the ability to replace momentum in factor models, and robustness among large and liquid coins and after trading costs. Our reconstruction and code audit identify issues in information timing, technical indicators, forecast coverage, portfolio formation, and return accounting. We first correct future-information use and related settlement rules, then implement broader corrections to indicators, forecast availability, and portfolio accounting. Without additional coin exclusions, the first stage changes CTREND's mean weekly gross return from 4.07% to 4.25%, annualized Sharpe ratio from 2.08 to 1.48, and weekly alpha relative to published market, size, and momentum factors from 2.77% to 3.06%. After comprehensive correction, these become 0.05%, 0.01, and -1.38%; the alpha is no longer significant. Replacing momentum with CTREND yields an advantage in maximum squared Sharpe ratios of 0.058 in the original implementation, 0.024 after the first-stage correction, and -0.0185 after comprehensive correction, with a statistically significant overall decline. Mean weekly gross returns for the largest 100 coins change from 2.93% to 3.99% to 1.00%, while mean weekly net returns for the full long-short portfolio under the lowest assumed trading costs change from 2.93% to 3.35% to -0.87%. These findings show that reported performance and factor-substitution gains depend materially on implementation and sample composition. We distinguish predictive information, an independent return premium, and the incremental value of replacing an existing factor as separate economic claims.

0.43

MaterialVellum: Steering Generative AI by Leveraging Expressive Material Affordances to Form Design Constraints

cs.HC ·2026-11-01 ·Chu Li, Nicolai Marquardt, Hugo Romat, Ken Hinckley, Nathalie Riche +2 more
top venue
Abstract

We present MaterialVellum, an interaction approach for steering generative AI through material-inspired manipulation of design constraints. In urban design, for instance, zoning bounds can be defined by the stretch of elastic bands, building density redistributed by molding clay, and functional relationships organized by the pull of magnets. MaterialVellum departs from current interfaces that rely on natural language prompts, and instead leverages users' existing knowledge of physical materials to represent and manipulate design constraints. We introduce a framework that maps material affordances---such as elasticity, malleability, or attraction---to spatial, relational, and proportional constraints that underpin design. These mappings are operationalized in user interfaces through familiar direct manipulation multi-touch gestures, including stretching, sculpting, and clustering, whose effects feed as control parameters into generative AI pipelines. We demonstrate this approach through technology probes, including an urban design, a flower arrangement design, and a graphical design scenario.A qualitative study (N=10) shows that material-based interactions improve accessibility of constraint expression, support exploratory workflows, and lower barriers for non-experts, while effectiveness depends on alignment with users' mental models.

0.43

Rubric-Based Assessment of Student Reports Using Small Language Models

cs.CY cs.CLcs.CR ·2026-11-01 ·Sandesh Kumar Kafle, Jan Vykopal, Valdemar Švábenský
publishedcode/data
Abstract · Full paper accepted for the 58th ACM Technical Symposium on Computer Science (SIGCSE TS 2027)

Language models can analyze and assess student texts using predefined rubrics and deliver structured feedback, reducing instructors' workload. However, existing solutions often rely on external services that are resource-intensive, costly, and raise privacy concerns, making them unsuitable for constrained environments. This study evaluates an automated assessment pipeline with three key aims: design for low-resource environments, privacy, and focus on a novel context in cybersecurity education. The pipeline integrates lightweight language models for textual evaluation and a vision-language model for image interpretation. Responses from 105 students were collected over two semesters of an undergraduate course during a cybersecurity exercise on incident response. Results across different hardware configurations show that small models in low-resource environments can achieve up to 97% accuracy in text-only and up to 94% in rich-text assessments. This rivals or even outperforms a large model while adding the benefits of locally hosted assessment. Our solution can help educators and practitioners who need grading assistance without replacing human oversight, especially in environments where cloud-based solutions may be impractical. This study contributes an open-source, locally deployable AI-assisted scoring pipeline that runs on affordable hardware and can be extended to various types of student reports in computing education. As a result, it makes automated assessment accessible to a broad range of educators and institutions.

0.41

AIDA: An AI Outage Registry with Evidence from On-Chain MEV

cs.CE q-fin.TR ·2026-11-01 ·Wenbin Wu
Abstract · 27 pages, 7 figures, 5 tables

Generative AI coding tools have lowered the cost of writing trading software, but no systematic dataset records when these services fail. We introduce the AI Incident and Disruption Archive (AIDA), a multi-vendor registry of 5,868 deduplicated outage incidents across 30 providers from 2025 to 2026, with foundation-model incidents verified against primary status-page or postmortem URLs. Pairing AIDA with 5,041,034 verified maximal extractable value (MEV) transactions by 8,516 Ethereum addresses, their contract creations, and a 22-month Base Layer-2 panel of 1,099,593,191 swaps and 29,120 private executors, we establish two findings. First, entry into MEV expanded where code and public state suffice, as active arbitrage addresses quadrupled from late 2024 to late 2026 and liquidators grew by a third, while sandwich attackers, who need private order flow and builder access, declined. Execution nonetheless remained concentrated, with the top fifty addresses executing two thirds of verified trades. Second, AI outages stall new trading code while deployed bots keep trading. On Ethereum, addresses releasing a private executor fall to 0.67 on the day of an eight-hour disruption (0.68 for executor creations during OpenAI outage hours) and to 0.66 at lag two before rebounding on day three; on Base, operator top-level creations fall to 0.76 during OpenAI hours and executor-deploying operators to 0.88 at lag two. Non-trading creations, automated factory clones, and live arbitrage and sandwich execution across both chains remain unchanged. Cheaper code thus lowers the barrier to building MEV contracts, while private order flow, latency and capital still determine who wins execution.

0.38

Gamification Box: Automated Personalized Rewards for Improved Task Experience

cs.HC ·2026-11-01 ·Ali Utku Ozaslan, Sadra Sadraddini, Deniz Marti
published
Abstract · Presented and published in Meaningful Play, 2026

We introduce a gamified rewarding system, called "Gamification Box", and conduct a user study to examine its effects on task experience and performance. Gamification Box provides an interface for automated rewards that are intrinsically enjoyable activities, such as watching short video clips. In a controlled user experiment, 118 participants completed a short-term memory task involving meaningless syllables under either a gamified reward condition or a non-rewarded control condition. Participants in the gamified condition reported significantly greater perceived task enjoyment than those in the control group. Importantly, this increase in perceived enjoyment did not come at the cost of performance: no significant differences were observed in free recall or recognition accuracy between conditions. These findings suggest that while gamified rewards may not directly enhance immediate performance, they function as a motivational regulator, thereby improving task enjoyment. The primary benefit of such systems may therefore lie in supporting sustained effort over time, where cumulative performance gains are more likely to emerge.

0.36

Harassment on TikTok: A Mixed-Methods Analysis of Video Content and Psycholinguistic Comment Stances

cs.HC cs.CY ·2026-11-01 ·Afsaneh Razi, Tyler Chang, Michal Monselise, Joseph J. Trybala, Asta Zelenkauskaite
published
Abstract · 34 pages

Despite extensive research on harassment, how it manifests on short--form video platforms such as TikTok---and how it shapes user interactions---remains underexplored, though these dynamics are critical for designing effective safety strategies for online communities. We present a mixed-methods study of 190 TikTok videos tagged with hashtag harassment and their 66,337 comments. Our qualitative analysis of videos identifies harassment types (sexual, verbal, stalking, law enforcement, bullying, catcalling, and gaslighting) and their characteristics (e.g., location, target, video style, and theme), along with engagement patterns. Although sexual harassment was the most prevalent topic, it did not have the highest engagement; instead, stalking--related videos received the most views. Pairwise comparisons and regression--based analyses of comments' psycholinguistic features and stance (support, opposition, neutrality) revealed differences in emotional tone, social orientation, and cognitive categories across harassment types. Comments on sexual harassment frequently express solidarity with victims, whereas discussions of law enforcement harassment emphasize institutional critique. We provide implications of our findings, demonstrating that harassment on TikTok is multifaceted, shaped by content, context, and community norms.

0.35

Learning structural response operators across varying geometries and loads with global and local geometry encoding

cs.CE ·2026-11-01 ·Xin Liu, Wuwen Yao
benchmark
Abstract · 27 pages, 20 figures, 7 tables; v2: Removed the template footer; scientific content unchanged

Predicting mechanical responses across complex geometries and varying loading conditions is central to simulation-driven engineering, yet capturing the influence of geometric boundaries remains challenging for data-driven models. Here we propose a Load- and Geometry-adaptive Deep Operator Network (LG-DON) that integrates global geometric representations with local, boundary-aware information to predict structural responses. By distinguishing distances to different boundary categories rather than relying on the nearest boundary alone, the framework captures spatially varying geometric influences that conventional distance-based representations may overlook. Across benchmark problems involving diverse non-convex geometries, LG-DON reduces the mean relative L2 error from 12.17% for a modified MIONet baseline to 7.41%. Controlled ablation studies further demonstrate the advantages of category-resolved boundary distances and show that combining global and local geometric representations outperforms either representation alone in 15 of 18 evaluated configurations. These findings highlight the value of integrating complementary geometric information for accurate operator learning across complex structural domains.

0.34

Measuring "Success" In the Transition from Pre- to Post-LLM Conversational Agents through the Lens of Interaction & Explanations for Older Adults

cs.HC ·2026-11-01 ·Tamara Zubatiy, Niharika Mathur, Reina Chan, Srijan Pokharel, Rosemarie Santa González +3 more
Abstract · Accepted by GROUP Companion '27

Conversational agents (CAs) for older adults have moved through at least two distinct eras, split by the emergence of large language models (LLMs) at the end of 2022. The first era was defined by commercial smart speakers such as the original Google Home and Amazon Alexa, built on rigid, rule-based dialogue. Interactions were single-turn, user-initiated and "successful" if and only if a command matched one of a finite set of pre-templated intents. The current era is marked by LLM-powered systems capable of open-ended, multi-turn conversation, with capacity for personalization and foundations for more proactive interaction. This shift has not only transformed what CAs can do for users, but also changed what researchers can measure when it comes to understanding "success" of user interactions with CAs. In this case study, we reflect on over 8 years of research spanning more than 100 participants across multiple field sites to trace how our own working definitions of "success" have evolved. We reflect on two completed bodies of work with pre-LLM CAs and an in-progress study with a post-LLM CA. We discuss how our definition and measurement of success have evolved during this transition and also ask open questions to the community about how to optimally measure "success" in the post-LLM CA era. This research is deeply applied and subject to the messiness of real-world deployments. Our intended contributions are deliberately practical, a combination of methodological considerations and reflection questions to support researchers engaging with this new wave of technologies.

0.33

Endless VR Walk: Dynamic Mapping of Limited Physical Spaces into Infinite Linear Paths

cs.HC ·2026-11-01 ·Sofia Martinez, Reagan Haddix, Miles Haase, Angelos Barmpoutis
published
Abstract · 2026 ACM SIGGRAPH Conference on Motion, Interaction and Games

Natural locomotion in virtual reality (VR) can provide a more immersive and engaging experience, but the physical space required for unrestricted walking is a significant limitation. This paper presents Endless VR Walk, a system for dynamically mapping a limited physical space into an effectively infinite linear virtual path. The approach uses a recursive path-generation algorithm to construct virtual blocks in real time as the user walks, while constraining the path to the available physical area and preventing spatial overlap. We evaluate path variability under different physical-space constraints and identify the smallest space that enables diverse path configurations. The resulting framework provides a platform for applications benefiting from extended walking, including physical therapy, rehabilitation, and fitness, as well as adaptations of traditional platformer and endless-runner games in VR.

0.32

Agent-Based Disruption Simulation on an Edge-Level Product-Annotated Supply Network of the Automotive Industry

cs.SI econ.GN ·2026-11-01 ·Benjamin Rolf, Hiroyasu Inoue, Sebastian Lang
Abstract · 46 pages, 10 figures, 10 tables. Submitted to Applied Network Science

Disruptions spread through supply networks when customers cannot replace the products they lose. We ask how portfolio-overlap and product-specific rules for selecting candidate suppliers change cascades when customers search for new suppliers and form new supply edges. From the MarkLines database, we construct a product-annotated network of about 77,000 firms and 197,000 supplier-customer edges in the global automotive industry. We extend an agent-based disruption model with edge-level products, a product-specific substitution rule based on the semantic similarity of product names, and calibrated capacities and inventories. Only 6% of the edges carry observed products. We impute the others and test the imputation against the annotated edges. We run 119,000 simulations. With all other mechanics fixed, the product-specific rule makes the cascades after a hub failure four to five times larger than the portfolio-overlap rule of the original model. Under alternative imputations and similarity thresholds, the factor ranges from 1.2 to 18. Within the simulations, substitutability also shapes which suppliers are critical. Network topology largely determines which firms a shock reaches, and substitutability determines how many of them fail. Of the 50 most damaging suppliers under random disruptions, topology alone identifies 20 and an index that adds product substitutability identifies 32. The index carries over without refitting to a regional scenario modelled on the 2011 Tohoku Earthquake. Supply network risk assessment should account for product-level substitutability and therefore needs product data on supply relationships.

0.31

Improving Entrepreneurial Skills by Practicing with AI

cs.CY ·2026-11-01 ·Amir Globerson, Anisha Choudhury, Anna Iurchenko, Avinatan Hassidim, Cathy Cheung +19 more
Abstract

Durable skills like critical thinking, creativity, and collaboration are essential for innovation, yet teaching them remains difficult. In this work, we focus on the entrepreneurial domain, where these skills and others are required, and ask whether AI can be used as a tool for improving entrepreneurial proficiency. As a first step towards this challenge, we present an extension of Vantage, a platform developed at Google Research, adapted to serve as an interactive practice environment for entrepreneurship. Vantage pairs students with simulated AI peers engineered to navigate realistic business scenarios and challenge learners to exercise key skills. In a randomized controlled trial (RCT) embedded in summer entrepreneurship programs run by The Network for Teaching Entrepreneurship (NFTE), students using Vantage scored significantly higher on their Capstone Business Plans than control students with access to standard AI tools. These preliminary findings suggest that orchestrated AI peers can meaningfully support short-term performance on tasks requiring complex, durable skills.

0.30

The Vibe Paradox: What EEG Reveals About Your Brain When Vibe Coding

cs.HC ·2026-11-01 ·Olivia Schutz, Rudrajit Choudhuri, Yunhan Qiao, Fatima A. Moussaoui, Julia Russo +2 more
Abstract · 12 pages, 5 figures, 1 table. Olivia Schutz and Rudrajit Choudhuri are co-first authors and have contributed equally

Background: Much of the enthusiasm for vibe coding rests on two intuitive beliefs: (1) that developers stay invested even when they no longer write the code, and (2) that delegating more of the work to AI lowers their cognitive load. Objective: Our goal was to investigate whether developers' neurophysiological states and interactions with AI are consistent with these beliefs when they work with agentic vs. non-agentic AI on their own project vs. a legacy codebase. Method: We conducted an exploratory within-subjects crossover study with 11 participants across 22 sessions. Each participant used GitHub Copilot Chat in one session and Cursor Agents in the other, working on their own project and a legacy codebase, with tool-codebase pairings randomly assigned and session order counterbalanced. We recorded participants' EEG, on-screen activity, and prompts, and analyzed 7,924 EEG data points, 984 activities, and 141 prompts. Results: Our findings contradict both beliefs. With the agent, participants remained indifferent to task outcomes, even on their own projects, and cognitive load was higher than with non-agentic AI. Furthermore, their ways of interacting shifted: with non-agentic AI, participants asked how the code worked and read it to verify changes; with the agent, they described desired app behavior and checked results in the running app. Conclusion: Our study reveals a vibe paradox: delegating to an agent left participants motivationally disinterested, even on their own projects, without relieving cognitive load, while coinciding with increased trust in AI. The community should treat this paradox as a first-class concern if developers and their AI agents are to go far together, and not just fast.

0.28

Long-term yields of short-rate models as Feynman-Kac ground states: a variational quantum eigensolver with single-setting energy estimation on two NISQ platforms

cs.CE quant-phq-fin.CPq-fin.MF ·2026-11-01 ·Doyeol "David" Ahn
Abstract

We show that the long-term yield of a short-rate model is the ground-state energy of a Schrodinger operator obtained from the Feynman-Kac bond-pricing equation, so that it can be computed by a variational quantum eigensolver (VQE). We treat the Vasicek, Cox-Ingersoll-Ross and Black-Karasinski models; the last has no closed form. Calibrated monthly to the U.S. Treasury curve from 1985 to 2026, the long-term yield follows the 30-year yield with an in-sample correlation of 0.98. The Hamiltonian is stoquastic, so its ground state is positive and the energy can be estimated from computational-basis samples with a single measurement setting. On ibm_cleveland, gate noise appears mainly as bit flips of the high-order qubits. A bit-flip deconvolution with a tail set fixed by a WKB criterion, combined with readout mitigation and zero-noise extrapolation, gives four-qubit estimates within 33-125 bp of the grid ground-state energies for all three models, and within 23-67 bp for Treasury-calibrated parameters. On IonQ Forte Enterprise the same analysis reduces the error in the same proportions. On the same circuits, a Pauli-measurement VQE deviates by 791-1,129 bp, so the single-setting estimator gives about ten times smaller errors with one setting instead of nine. An on-device VQE converges from 4,939 bp to 130 bp.

0.27

Using LLM Agents to Evaluate Prosocial Interventions on Social Media Platforms

cs.SI cs.CL ·2026-11-01 ·Lucen Li, Erin Wertz, Muhammad Haroon, Anshuman Chhabra, Magdalena Wojcieszak
Abstract

Social media platforms are a primary source of information for many users. Although research explores various interventions to promote prosocial content or demote harmful content on platforms, there are few systematic methods for evaluating competing intervention designs before conducting costly human-subject experiments. We develop a simulation-based evaluation pipeline built on the OASIS framework. We generate 13,835 LLM-based agents, each with a unique profile. Over 20 time steps, we expose them to real-world Twitter/X posts and let them comment, like, repost, or follow. We use those agents to systematically evaluate three interventions that increase exposure to verified news: adding content, upranking content, and celebrity source cues. All interventions significantly increase news engagement, with upranking news to prominent feed positions producing the largest effects. These effects hold across agents' gender, age, and education levels. We validate our injection treatment results against a comparable human-subject study and find strong directional alignment and similar dose-normalized efficiency. Our results suggest that LLM-based agent simulations can support early-stage evaluation of platform interventions before human-subject deployment, and upranking may be more effective than news volume or source.

0.26

When the Text Is Silent: Memory Shortcuts and the Validity of LLM Annotation under Evidence Insufficiency

cs.HC cs.CL ·2026-11-01 ·Wenchang Wang, Yue Song
Abstract · 15 pages, 5 figures

Large language models (LLMs) are increasingly used to annotate text for social science research, content moderation, and training data. Their broad knowledge of real-world entities raises a concern: when labeling a text, do they rely on its evidence or substitute what they already know? We call this behavior a memory shortcut and examine when it emerges, what amplifies it, and how to mitigate it. Using 995 widely recognized public figures and four matched control names per figure, we construct counterfactual texts that separate textual evidence from prior knowledge, and evaluate three commercial LLMs on 121,594 prompts each across five stages. We find a striking asymmetry. Although all models recognize nearly every figure, none follows memory over the text when the two conflict. Shortcuts instead emerge when evidence is absent: for one model, famous names raise the rate of remembered occupations, ages, and countries by 17.0, 14.9, and 6.8 percentage points relative to synthetic names. These effects occur only for recognized figures, and 98.6% of the resulting labels are factually correct about the namesake while none is supported by the text. Evaluation threats and answer-contingent rewards amplify shortcuts by 8.7 and 12.4 points. Famous names also raise leadership ratings of sparsely documented applicants by 0.22 and 0.28 points in two models. A namesake reminder and name masking each eliminate the shortcuts, with trade-offs in gender annotation. LLM annotations can thus be factually correct yet methodologically invalid. Trustworthy LLM annotation requires audits that test evidence absence, account for incentives in annotation instructions, and evaluate the attribute-specific trade-offs of mitigation.

0.25

User Engagement in Visualization: A Scoping Review of 243 Research Articles

cs.HC ·2026-11-01 ·Mara Solen, Clara-Marie Barth, Heather O'Brien, Tamara Munzner
Abstract

User engagement is well studied in other domains. Work to date has proposed conceptual models and definitions, approaches to measurement, and factors that influence engagement in digital interactions. However, of the articles in the visualization literature that refer to engagement, few of them make evidence-based, generalizable claims, and research that does is challenging to locate due to inconsistent reporting and often lacks connection to the broader engagement literature. To address these issues and promote the consideration of the broader engagement literature, we conduct a comprehensive scoping review of visualization literature and engagement. Using scoping review methodology, we identify and analyze 243 articles that make claims about engagement with visualizations. We coded the studies conducted, literature cited, definitions provided, and claims made in these articles and analyze the codes to evaluate the quality of the work in the field and to synthesize existing findings. We find that, despite frequent references to engagement and apparent interest in the topic, the overall quality of visualization engagement research is poor, often with weak evidence, inconsistent conceptualizations, and infrequent citation of existing models, methods, or claims. We report the evidence-based claims about what engages people with visualization, including the factors of animation, interaction, and immersion, additionally describing both corroboration and apparent contradictions between these claims. Finally, we propose the COMFI framework and mnemonic for reporting important information in future engagement studies.

0.25

No Note Is an Island: Probing LLMs' Understanding of Crowdsourced Fact-Checking Environments

cs.SI ·2026-11-01 ·Weijian Fan, Chichen Lin, Xiao Han, Jiaying Wu
Abstract

Large language models (LLMs) increasingly assist crowdsourced fact-checking, yet evaluations focus on individual annotations. A correction's reach and reception depend on visibility, audience responses, and platform policies. We introduce correction context understanding: assessing corrective content, interpreting audience responses, and reasoning about platform dynamics within a shared environment. We build NoteHabitat from 36 months of Weibo Annotations data, linking 102,994 posts (42,853 annotations), 431,407 comments, and five policy events. Evaluating diverse model families across seven tasks spanning content, interaction, and management reveals three gaps. Annotation quality correlates only weakly with engagement ($ρ=0.068$), while four zero-shot LLMs systematically underrate quality relative to human judgments, especially evidence support and objectivity. Models also miss toxic comments supporting corrections (36.4% average recall across ten models). At the platform level, LLMs partially predict policy-associated outcome directions, but LLM-guided simulations poorly reproduce daily activity. Cross-group analyses expose blind spots tied to contribution types and reader stances that aggregate scores conceal. Reliable fact-checking AI must understand the note, the conversation around it, and the platform in which both unfold.

0.25

Cross-Scale Spatial Reasoning: A Learning-Centered Framework for Designing Immersive Learning Environments

cs.HC ·2026-11-01 ·Dishita Turakhia
Abstract

Understanding complex scientific and engineering systems often requires learners to connect structures, processes, and behaviors across different spatial scales and levels of organization. Although research on spatial cognition, scale, multiple representations, and systems thinking has examined aspects of this challenge, these perspectives do not always provide explicit guidance for designing learning activities that require learners to integrate information across levels. Immersive technologies offer opportunities to explore otherwise inaccessible structures, coordinate representations, and investigate system behavior, but their educational value depends on the reasoning they support. This paper proposes Cross-Scale Spatial Reasoning (CSSR) as an integrative framework for understanding how learners establish and use spatial and structural relationships across scales or organizational levels. We describe five interrelated processes: scale navigation, cross-scale relational mapping, representational coordination, dynamic and causal reasoning, and consequence-oriented reasoning. Building on these processes, we develop a learning-centered framework that connects reasoning objectives to immersive design mechanisms, instructional scaffolding, and assessment. Illustrative applications in biology, semiconductor design, manufacturing, and architecture show how the framework can inform domain-specific learning activities. We also propose testable research propositions and approaches for evaluating cross-scale understanding, instructional effectiveness, and transfer. Rather than treating immersion as an educational outcome in itself, CSSR provides a basis for designing and studying learning environments that help learners explain how complex systems work across levels.

0.24

Supporting Orthographic Recall of Chinese Characters through Self-Generated Compositional Mnemonic Stories: A Mixed-Methods Pilot Study

cs.HC ·2026-11-01 ·Aaron Friedrich Kurz
Abstract · TCLT13: The 13th International Conference and Workshops on Technology and Chinese Language Teaching, Oct 2026, The University of Hong Kong

Mnemonic studies on Chinese characters have mostly tested provided mnemonics against writing outcomes and self-generated mnemonics against meaning recall; whether writing one's own component-based story helps learners write the character has received little attention. I present Nuome, a tool that shows a character's component decomposition next to an editor in which learners write their own story, with live checks on whether the story covers every component and the character's meaning. In a within-subjects pilot, eight L2 learners each learned four self-chosen characters with the story builder and four with a read-only panel showing the same decomposition, with identical writing practice. Descriptively, partial-credit written-recall scores favoured the story builder by 18.8 percentage points immediately (N = 8) and 17.9 points after two days (N = 7); mean study time per character was 1.69 times as long (n = 6); usability was high (SUS 88.1). Interviews yielded three themes: participants felt that composing a story made them attend to components; knowing the components is not the same as being able to write them; and characters differ in how well they suit a story.

0.23

Capable but Not Scalable: Requirements and Interaction Costs of Building Multi-View CAD Datasets for Design Research

cs.HC ·2026-11-01 ·Danyal Sarfraz, SeEun Lee, Ulugbek Vahobjon Ugli Ismatullaev, JoungHyun Kim, Abdurakhimov Donierbek +1 more
Abstract

Transforming CAD models into consistent multi-view image datasets is increasingly important for AI-driven product design research, yet existing workflows can impose substantial burdens on researchers without programming expertise. We examine this challenge from a human-computer interaction (HCI) perspective by analyzing ten reported rendering practices and deriving fourteen requirements spanning dataset quality, visual consistency, and user interaction. We then assess eleven development- and design-oriented tools against an illustrative task involving STEP models, interpreting potential interaction costs through the cognitive dimensions framework. Our analysis reveals a gap between technical capability and practical scalability: programming-based frameworks offer flexible automation but require coding expertise, while graphical design tools provide accessible functionality yet retain repetitive per-product operations. No assessed workflow combines a low programming threshold, low repetition viscosity, and flexible protocol configuration. Based on these findings, we propose design directions including reusable rendering presets, collection-wide previews and corrections, mixed-initiative orientation assignment, and automatic provenance recording. This work reframes CAD dataset generation as an end-user interaction design problem and establishes requirements for more accessible, scalable dataset creation tools.

0.19

Theoretical relations between basic centrality measures in real-world networks

cs.SI math.CO ·2026-11-01 ·Mikhail Tuzhilin
Abstract

We establish new relationships among several fundamental measures of real-world networks, including the average clustering coefficient, the global clustering coefficient, the average shortest-path length and degree centrality, closeness centrality, betweenness centrality and ksi-centrality. We further develop the theory of $j$-neighborhood centrality, which encompasses degree centrality and ksi-centrality as special cases.

0.14

Interface Quilt: Modifying Applications with Computer Use Agents and Generative UI

cs.HC ·2026-11-01 ·Leping Qiu, Jo Vermeulen, Tim Zindulka, George Fitzmaurice, Fraser Anderson
Abstract · Preprint, under review

Modifying an existing application is difficult when it does not provide programmatic access for that change. Recent advances in computer-use agents (CUAs) enable richer understanding and operation of existing interfaces, while generative UI (GenUI) makes it practical to generate new ones. We examine how these advances expand software modification through a design space of what can be changed and a framework for how modifications can be implemented. We then introduce Interface Quilt, a system that combines CUAs and GenUI to generate interface patches, which are new interfaces displayed over an existing application to modify its presentation, interaction, and capabilities. Through demonstrations across a variety of applications, we show that interface patches support a range of modifications without source code, APIs, or accessibility data.

0.08