CS Paper Radar

Weekly top 20

2026-W41 Oct 05 – Oct 11, 2026

Ranked from 5,698 papers submitted on 7 days. Each research area gets slots in proportion to its share of this week's papers (at least one), filled by score; near-duplicates are removed. Why

Score = weighted mean ofCitations 20%Upvotes 10%Code stars 10%Venue 15%Code/data 10%Novelty 15%Momentum 15%Substance 5%how it works
1

Teaching a Minimalist Machine to Discover Recursive Programs for Arithmetic

cs.AI ·2026-10-05 ·Dominik Magiera, Christiane Wiebel-Herboth, Frank Jäkel
publishedcode/data
Abstract · 15 pages, 5 figures. Accepted for oral presentation at the 6th International Joint Conference on Learning and Reasoning 2026. Code: https://…

Humans can often acquire and synthesize complex, recursive concepts from minimal experience. Leveraging cognitive insights, we propose the Minimalist Machine, a framework for inductive program synthesis designed to model such conceptual learning. The system uses a compact relational subset of Prolog: Programs are searched within a fixed schema of body-free facts and two-body conjunctive Horn clauses. Recursion is not defined by a dedicated metarule. Instead, it emerges when a target predicate is reused inside the body of a learned clause. Inspired by a primary school curriculum, the model is taught through a human-curated, sequential introduction of new concepts in arithmetic. Starting from initially empty knowledge base, it first acquires simple structural predicates, then successor-based state transformations, and finally recursive programs for addition, subtraction, multiplication, and division. Ultimately, this approach yields the fully transparent, inductive reasoning trace necessary for human-like conceptual learning.

0.79
2

RES-MR: Risk-Aware Reasoning for Explainable and Safe Medication Recommendation

cs.LG cs.AIcs.IR ·2026-10-10 ·Cong Wang, Jin Li, Shoujin Wang, Yishuo Li, Huilin Gu +1 more
top venuebenchmarkcode/datahighly cited
Abstract · 12 pages

Medication recommendations seek to deliver personalized drug combinations based on patients' clinical information, where ensuring transparent decision-making and personalized safety are of paramount importance. However, existing approaches predominantly operate as ``black boxes'', offering limited interpretability and hindering clinician trust. Furthermore, they typically rely on fixed, patient-agnostic safety constraints, neglecting the heterogeneity of risk tolerance across individuals, which poses significant risks to vulnerable populations. To address these gaps, we propose \textbf{RES-MR}, a novel Risk-aware Reasoning framework for Explainable and Safe Medication Recommendation using LLMs. RES-MR follows a two-stage training paradigm: (1) an explainable clinical reasoning distillation stage to elicit diagnostic capabilities by fine-tuning on synthesized reasoning trajectories, grounded in patient-specific knowledge from graph retrieval; and (2) a risk-aware policy optimization stage to dynamically balance therapeutic efficacy with personalized safety. This stage employs disentangled drug factors to construct patient-specific reference prototypes that capture desired therapeutic drug features and risk features to avoid. These prototypes enable risk boundary calibration and guide the model toward safe recommendations via reward shaping and clipping. Extensive experiments on benchmark datasets demonstrate that RES-MR significantly outperforms state-of-the-art baselines in terms of accuracy, safety, and explainability. Our code and data are available at: https://github.com/wangcong2001/RES-MR.

0.78
3

Scalable Minimal-Change Learning for Controllable Image Editing

cs.CV cs.AI ·2026-10-05 ·Shuo Chen, Fengming Huang, Yu Yao, Mingming Gong, Tongliang Liu
top venuecode/datahot direction
Abstract · 29 pages, 2 figures. Accepted at NeurIPS 2026. Revised version with additional off-target, reward and SFT control, human evaluation, and Omn…

Image editing should change only the attributes specified by an instruction while preserving everything else, yet current methods often make unintended changes. We treat this minimal-change principle as an optimization objective for instruction-based editing. Latent L1 regularization is a poor proxy for output locality in modern nonlinear generators and often requires supervision unavailable at scale. We instead optimize edit outcomes with reinforcement learning. An agentic vision-language reward model audits each source image, instruction, and edited image for two failure types: unimplemented requested changes and unintended changes. A group-level rubric merges and verifies these issues to provide consistent rewards across candidate edits without per-instruction human annotations. On FLUX.1 Kontext-dev, ARRO raises average EditScore from 5.21 to 5.88 across MinEval, MagicBrush, AnyBench, and Emu-Edit. On 600 evaluation examples, it reduces off-target pixel change by 8.4% relative to the base editor. Reward and SFT controls, blinded human evaluations, and transfer to OmniGen2 provide complementary evidence. Code: https://github.com/Showwwwwwwww/ARRO

0.77
4

AliO: Output Alignment Matters in Long-Term Time Series Forecasing

cs.AI ·2026-10-08 ·Kwangryeol Park, Jaeho Kim, Seulki Lee
top venue
Abstract · NeurIPS 2025. 46 pages

Long-term Time Series Forecasting (LTSF) tasks, which leverage the current data sequence as input to predict the future sequence, have become increasingly crucial in real-world applications such as weather forecasting and planning of electricity consumption. However, state-of-the-art LTSF models often fail to achieve prediction output alignment for the same timestamps across lagged input sequences. Instead, these models exhibit low output alignment, resulting in fluctuation in prediction outputs for the same timestamps, undermining the model's reliability. To address this, we propose AliO (Align Outputs), a novel approach designed to improve the output alignment of LTSF models by reducing the discrepancies between prediction outputs for the same timestamps in both the time and frequency domains. To measure output alignment, we introduce a new metric, TAM (Time Alignment Metric), which quantifies the alignment between prediction outputs, whereas existing metrics such as MSE only capture the distance between prediction outputs and ground truths. Experimental results show that AliO effectively improves the output alignment, i.e., up to 58.2% in TAM, while maintaining or enhancing the forecasting performance (up to 27.5%). This improved output alignment increases the reliability of the LTSF models, making them more applicable in real-world scenarios.

0.77
5

PerSpectron: Detecting Invariant Footprints of Microarchitectural Attacks with Perceptron

cs.CR ·2026-10-06 ·Samira Mirbagher-Ajorpaz, Gilles Pokam, Esmaeil Mohammadian-Koruyeh, Elba Garza, Nael Abu-Ghazaleh +1 more
top venuehot directionhighly cited
Abstract · Author manuscript of MICRO 2020. 14 manuscript pages plus publication-notice page

Detecting microarchitectural attacks is critical given their proliferation in recent years. Many of these attacks exhibit intrinsic behaviors essential to the nature of their operation, such as creating contention or misspeculation. This study systematically investigates the microarchitectural footprints of hardware-based attacks and shows how they can be detected and classified using an efficient hardware predictor. We present a methodology to use correlated microarchitectural statistics to design a hardware-based neural predictor capable of detecting and classifying microarchitectural attacks before data is leaked. Once a potential attack is detected, it can be proactively mitigated by triggering appropriate countermeasures. Our hardware-based detector, PerSpectron, uses perceptron learning to identify and classify attacks. Perceptron-based prediction has been successfully used in branch prediction and other hardware-based applications. PerSpectron has minimal performance overhead. The statistics being monitored have similar overhead to already existing performance monitoring counters. Additionally, PerSpectron operates outside the processor's critical paths, offering security without added computation delay. Our system achieves a usable detection rate for detecting attacks such as SpectreV1, SpectreV2, SpectreRSB, Meltdown, breakingKSLR, Flush+Flush, Flush+Reload, Prime+Probe as well as cache-attack calibration programs. We also believe that the large number of diverse microarchitectural features offers both evasion resilience and interpretability---features not present in previous hardware security detectors. We detect these attacks early enough to avoid any data leakage, unlike previous work that triggers countermeasures only after data has been exposed.

0.76
6

SquidAgent: Parallelize Wisely, Coordinate Efficiently

cs.AI cs.CLcs.LG ·2026-10-06 ·Yexiong Lin, Shanshan Ye, Yu Yao, Zhen Fang, Bo Han +1 more
top venuehighly cited
Abstract · Accepted at NeurIPS 2026. 37 pages, including appendices

LLM-based agents solve complex multi-step tasks, but sequential execution incurs substantial latency. In principle, parallelizing work across multiple agents should yield near-linear speedups. Yet existing parallel multi-agent systems often run slower than a single-agent baseline. We attribute this gap to two hidden costs that parallel execution incurs but a serial agent avoids. First, there is a re-exploration cost: redundant effort spent by parallel workers reconstructing context that the orchestrator already possesses, such as prior decisions, that would otherwise be inherited implicitly in a serial execution. Second, there is an alignment cost: the overhead required to reconcile inconsistencies across independently generated outputs. We thus derive a principled decision criterion: a layer should be parallelized only when its critical-path cost, plus re-exploration and alignment overheads, is lower than the corresponding serial cost. While this criterion is naturally expressed in wall-clock time, we observe that LLMs are poorly calibrated when asked to estimate task duration. To address this, we instead measure cost in predicted output tokens, which we empirically find LLMs can estimate substantially more reliably than wall-clock time. Building on this token-based criterion, we propose SquidAgent. It estimates all token budgets in a single planning step, forks each worker directly from the orchestrator's session to eliminate re-exploration cost, and replaces post-hoc reconciliation with a pre-generated shared convention block that converts alignment into a bounded upfront cost. A deterministic scheduler then applies the criterion layer by layer. Empirically, SquidAgent achieves a 2.2$\times$ mean throughput improvement and a 2.6$\times$ mean wall-time speedup over Claude Code, and a 2.0$\times$ throughput improvement over the strongest multi-agent baseline.

0.74
7

HoldSpec: A Machine-Checked Model and Conformance Suite for the Payment Authorization Hold Lifecycle

cs.SE ·2026-10-10 ·Abhishek Sharma
code/data
Abstract · 8 pages, 3 figures, 6 tables. Code, TLA+ specification, generated conformance suite, mock providers, and reproducibility materials: https://…

When a merchant authorizes a card and captures the funds later, the money sits in a state that neither party fully controls: held on the cardholder's account, not yet taken, and due to lapse on a deadline set by the card network. What a merchant may do to that hold, capture part of it, capture again, void it, raise it, let it expire, is business logic that every payment service provider implements slightly differently and none of them specifies formally. We give the lifecycle a state machine in TLA+, parameterized by a provider capability profile, and check eight safety invariants and three liveness properties over 6 profiles, five of them built from Stripe's and Adyen's published documentation. From the same machine we generate a conformance suite that runs black-box against a provider API. Against 61 seeded defects that are distinguishishable from a correct implementation, the generated suite detects 100%, where random sequences with the same call budget reach 56% and a hand-written integration suite reaches 69%; random sequences given ten times the budget still reach only 67%. Running the same suite between two providers turns the comparison into a search, and it returns short executable witnesses for divergences the two documentation sets imply but neither states, including one that breaks the common assumption that a single adapter interface can sit over both, because closing a hold and releasing its remainder is a capture on one provider and a cancel on the other. We report what we could not do: no provider sandbox credentials were available, so every result here concerns documented behavior, and the suite that would test the providers themselves is part of what we release.

0.74
8

Virtual model control for compliant reaching under uncertainties

cs.RO eess.SY ·2026-10-05 ·Yi Zhang, Daniel Larby, Fumiya Iida, Fulvio Forni
publishedhot directionhighly cited
Abstract

Virtual Model Control (VMC) is an approach to design a controller for force-controlled robots in complex uncertain environments. While this method was primarily investigated for legged robot locomotion in the past, it can be more generally applicable to other types of robotic systems. This paper investigates the VMC framework for reaching tasks in a force-controlled robotic arm. We propose six different approaches to designing virtual models in order to achieve reaching tasks in environments with obstacles and uncertainties. A force-controlled 8 degree-of-freedom humanoid robot was used to validate the proposed approach in the real world. We conducted three experiments to test the performance of VMC controllers in terms of predictability, sensitivity to external force, and adaptability against known and unknown obstacles. Experimental analyses show that, even though the proposed approach needs to sacrifice accuracy and trajectory optimality, it enables us to design complex reaching motions under uncertainties, in an intuitive and extendable manner.

0.74
9

RAPID: Proactive Rate Control for Enabling High Data Rate and Low Latency in 5G Edge Networks

cs.NI ·2026-10-09 ·Hamid Hassani, Sonia M. Heemstra de Groot, Ignas G. Niemegeers, Kishor C. Joshi, George Exarchakos
top venuehighly cited
Abstract

Many 6G use cases such as tactile Internet and eXtended reality applications would simultaneously require a high data rate and low latency wireless connectivity.It is increasingly being realized that the current transport protocols relying on packet loss and delay are insufficient to satisfy the data rate and latency requirements of such applications. To address this problem, we propose RAPID, a transport layer solution designed for private 5G edge networks, where the bottleneck is located at the base station. We show that by regulating the send rate to maintain a target physical resource block utilization, we can achieve high data rate, low latency communication. We design a simple yet effective policy that adapts to the dynamic nature of wireless channel conditions. In a multi-UE scenario with two RAPID flows and one TCP Cubic flow, our testbed results show that RAPID flows achieve significant improvements in one-way delay, i.e., 72x and 55x lower compared to TCP Cubic, with negligible packet loss at the cost of modest goodput reductions of 16% and 27%, respectively.

0.73
10

Demonstrating Arena 5.0: A Photorealistic ROS2 Simulation Framework for Developing and Benchmarking Social Navigation

cs.RO ·2026-10-08 ·Volodymyr Shcherbyna, Linh Kästner, Duc Anh Do, Hoang Tung, Huu Giang Nguyen +11 more
publishedbenchmarkcode/datahighly cited
Abstract · 12 pages, 10 figures. Published at Robotics: Science and Systems (RSS) 2025

Building upon the foundations laid by our previous work, this paper introduces Arena 5.0, the fifth iteration of our framework for robotics social navigation development and benchmarking. Arena 5.0 provides three main contributions: 1) The complete integration of NVIDIA Isaac Gym, enabling photorealistic simulations and more efficient training. It seamlessly incorporates Isaac Gym into the Arena platform, allowing the use of existing modules such as randomized environment generation, evaluation tools, ROS2 support, and the integration of planners, robot models, and APIs within Isaac Gym. 2) A comprehensive benchmark of state-of-the-art social navigation strategies, evaluated on a diverse set of generated and customized worlds and scenarios of varying difficulty levels. These benchmarks provide a detailed assessment of navigation planners using a wide range of social navigation metrics. 3) Extensive scenario generation and task planning modules for improved and customizable generation of social navigation scenarios, such as emergency and rescue situations. The platform's performance was evaluated by generating the aforementioned benchmark and through a comprehensive user study, demonstrating significant improvements in usability and efficiency compared to previous versions. Arena 5.0 is open source and available at https://github.com/Arena-Rosnav.

0.73
11

RollPlace: Improving Macro Placement via Monte Carlo Rollout Search

cs.AI ·2026-10-05 ·Qi Zhou, Guojun Liu, Guangzhi Qi, Ming Lu, Jiechu Liu +3 more
top venuebenchmarkhot direction
Abstract · 14 pages, 8 figures, 6 tables

The application of Reinforcement Learning (RL) in Electronic Design Automation (EDA), particularly for chip placement, has attracted considerable attention in recent years. While existing machine learning (ML)-based approaches have achieved notable progress, they predominantly focus on generating optimal layouts in a single attempt, often producing solutions that require subsequent refinement. To address this limitation, we propose RollPlace, a novel and generalized macro placement framework. RollPlace adopts a two-stage optimization strategy: generating initial placement solutions via machine learning methods or heuristic-based strategies, and refining these layouts efficiently by adjusting specific macros derived from the initial stage. This strategy circumvents the sequential generation constraints inherent in traditional RL-based placement methods. Furthermore, RollPlace seamlessly integrates Monte Carlo Tree Search (MCTS) to balance exploration and exploitation, and employs a rollout mechanism for efficient local search. Extensive experiments on the ISPD 2005 benchmark demonstrate that RollPlace outperforms state-of-the-art methods. Additionally, end-to-end experimental results based on OpenROAD across 19 benchmarks show that RollPlace excels in multiple metrics. The proposed framework offers a robust and scalable solution for addressing the growing complexity of modern chip design challenges.

0.73
12

WorldCast: Distributed Multiplayer World Models

cs.CV ·2026-10-08 ·Ziyang Ye, Junchao Huang, Evelyn Zhang, Zhihao Xie, Ruicheng Zhang +7 more
code/data
Abstract · 31 pages, 19 figures, 20 tables. Project page: https://ziyang-ye.github.io/WorldCast-Page

Multiplayer world models must generate independently controlled views with consistent representations of both players and their shared environment. Most existing approaches coordinate multiple players through joint multi-view generation, whose cost grows with each additional player. We present WorldCast, a distributed multiplayer world model in which each player runs a local client comprising a video generator and a state model. Using recorded player positions and map geometry during training, the state model estimates the player's position from generated video and control inputs. Clients exchange player states and project them into camera-aligned player state fields that guide where and how other players are rendered. Shared scene state enables clients to reuse one another's generated observations to maintain consistent scene appearance across views. Experiments on Counter-Strike 2 demonstrate WorldCast's consistency, real-time performance, and distributed scalability. The camera-aligned player state field improves player rendering rates by over an order of magnitude over joint-generation methods, while shared scene state improves visual consistency over whole rounds. Each client runs in real time and exchanges only player and scene states, enabling scalable multiplayer generation without a centralized computational bottleneck. Image quality remains stable over hour-long rollouts.

0.72
13

Local Citation Recommendation Without Graph Signals

cs.IR cs.CLcs.DL ·2026-10-11 ·Benjamin Basseri, Pavlos Protopapas, Kemeng Zhang, Alberto Accomazzi, Stephanie Jarmak +1 more
publisheddatasetcode/datahot direction
Abstract · Accepted at WASP 2026, co-located with AACL-IJCNLP 2026. 16 pages, 3 figures, 2 tables. Dataset: https://huggingface.co/datasets/adsabs/LCRC…

Resolving correct citations can be a time-consuming task for research writers. We address local citation recommendation (LCR), where the goal is to retrieve appropriate references for a given sentence. Web tools generally depend on the existing citation graph, which may bias results toward citation count or more recent publications. LCR research using full paper texts has been limited by the lack of publicly available data. We present a new dataset, the Local Citation Recommendation Corpus (LCRC), containing 14,735 scientific statements derived from astrophysics review papers, each with one or more citation targets whose full text is accessible via API. Using this dataset we systematically compare retrieval methods and develop a citation recommendation system that avoids citation graph bias while achieving 8.0% improvement in Recall@25 and 6.9% improvement in NDCG@25 over dense retrieval baselines without model fine-tuning. To support reproducibility, we make our code and dataset publicly available.

0.72
14

Drive3R: Block-wise Streaming Geometric Foundation Models for Autonomous Driving

cs.CV ·2026-10-10 ·Tianyi Yang, Haojun Yang, Jipeng Xiao, Xuesong Chen, Shaoshuai Shi +2 more
code/data
Abstract · 11 pages, 6 figures, 4 tables. Project page: https://silence-ai423.github.io/Drive3R/

High-fidelity closed-loop simulation for autonomous driving increasingly relies on dense multi-view 3D reconstruction. While recent feed-forward geometric foundation models have revolutionized per-scene optimization, applying them to continuous, long-horizon driving sequences is computationally intractable due to GPU memory limits and the quadratic complexity of global attention. Existing scaling strategies are fundamentally limited, as causal streaming suffers from compounding scale drift and trajectory deviation, while rigid "chunk-then-align" heuristics fragment global context and produce trajectory discontinuities. To address these limitations, we introduce Drive3R, a block-wise streaming geometric foundation model. Its block-wise causal attention resolves the scalability bottleneck of fully bidirectional attention while avoiding the loss of local future context in purely causal streaming. Its confidence-aware re-anchoring further mitigates long-range drift by detecting unreliable streaming states, re-initializing the KV cache, and stitching refreshed segments through SE(3) alignment. Extensive experiments demonstrate that Drive3R establishes a new state-of-the-art on nuScenes, KITTI, and Waymo. Specifically, on nuScenes, it improves over the recent DVGT-2 baseline from 0.1806 to 0.1081 in AbsRel, from 0.6750 to 0.5344 in point-cloud Accuracy, and from 1.5812 to 0.9092 in ATE.

0.71
15

3SUM Is Really Hard: A Real-to-Integer Reduction

cs.DS ·2026-10-07 ·Nick Fischer, Adam Polak, Jonas Schmidt
top venuetheory
Abstract · To appear in FOCS 2026

We show that if the 3SUM problem on integer-valued inputs can be solved in truly subquadratic time, then it can also be solved in truly subquadratic time on real-valued inputs. This answers an open problem posed by Chan, Vassilevska Williams, and Xu [STOC 2022], and constitutes the first such tight real-to-integer self-reduction in fine-grained complexity. Our proof relies on a surprising combination of the Frank--Tardos theorem from optimization with Freiman-type theorems from additive combinatorics.

0.71
16

CoDe-LoRA: Mitigating the Orthogonality Dilemma in Continual Learning of LLMs via Knowledge Consolidation and Decoupling

cs.AI cs.CLcs.LG ·2026-10-06 ·Maoqi Liu, Quan Fang, Yufei He
top venuecode/data
Abstract · Accepted to EMNLP 2026 (Main Conference)

Continual learning (CL) is essential for Large Language Models (LLMs) to sequentially adapt to evolving tasks. To mitigate catastrophic forgetting, recent advances implement low-rank adaptation with orthogonal projections (e.g., O-LoRA) to isolate task parameters. However, we reveal that such strict geometric constraints trigger an "Orthogonality Dilemma": rigid parameter isolation impedes the transfer and accumulation of shared representations across semantically related tasks. In this work, we propose a new replay-free method, called Consolidation and Decoupling LoRA (CoDe-LoRA), for CL of LLMs. CoDe-LoRA disentangles the learning process into Consolidating Universal Knowledge and Decoupling Task-Specific Knowledge. To achieve this, CoDe-LoRA leverages an adaptive null space projection mechanism and semantic routing to balance knowledge accumulation with task-specific adaptation. Experimental results across four backbones and three CL benchmarks show that CoDe-LoRA achieves the best average accuracy. Our code is available at https://github.com/Estrellajer/CoDe-LoRA.

0.70
17

MC-EvoBench: Auditing Experience-Driven Self-Improvement in Minecraft Agents

cs.RO cs.AI ·2026-10-09 ·Jingwei Song, Zhisheng Chen, Bingfan Zeng, Jing Tang, Yubo Hou +3 more
benchmarkcode/datahot direction
Abstract · 39 pages, 6 figures

A central challenge in embodied recursive self-improvement is enabling agents to convert interaction history into capabilities that persist beyond one world and transfer to later tasks. A saved memory or successful final action does not by itself establish that a reusable capability was acquired; evaluation must separate capability change from carried-over resources, unequal source exposure, and early execution failures. We construct MC-EvoBench as a controlled Minecraft benchmark with a fixed catalog of 19 registered evolution chains, including 12 primary evaluation chains and seven calibration or probe chains, together with three graph-aligned paired probe families for transfer. The benchmark represents each task as a task-state graph, defines stage-level environment verifiers, assigns matched source and target roles, resets the target world, and permits only a declared agent artifact to cross the source-target boundary. For each probe, no-prior, related, unrelated, and shuffled source conditions are evaluated under matched environment-step budgets and five seeds; we report TSR, WGP, and VMP and retain provenance records. The benchmark reveals a consistent direct-capability gap between controller and planner-backed variants, while related experience yields no stable transfer: it does not improve either tool probe, its small wood-species gain hinges on one seed, and all intervals include zero. Stage audits further show that early wood-collection and crafting bottlenecks often leave later capabilities unobserved. MC-EvoBench thus offers a controlled testbed for durable, transferable embodied capabilities. Code and data are available at https://github.com/xzw-ustc/mc-bench-open-source

0.70
18

DVLA-RL++: Dual-Level Vision-Language Alignment with Reinforcement Learning Gating for Few-Shot Learning

cs.CV cs.LG ·2026-10-08 ·Wenhao Li, Xianjing Meng, Qiangchang Wang, Zhongyi Han, Yilong Yin +1 more
top venuecode/datahighly cited
Abstract · This work has been submitted to the IEEE TPAMI for possible publication

Few-shot learning aims to recognize novel categories from limited labeled examples. Recent studies incorporate textual semantics to compensate for limited visual observations and improve class representations. However, high image-text agreement may reflect both intrinsic object properties and incidental context, making support prototypes susceptible to contextual contamination. To address this problem, we propose DVLA-RL++, which extends DVLA-RL with complementary semantic purification (CSP) and counterfactual reinforcement-learning gating (CRG). Specifically, CSP generates intrinsic and nuisance descriptions from labeled supports and compares their agreement with each support token. An ambiguity-dependent rejection margin guides sparse evidence allocation, while an intrinsic semantic anchor fills the unassigned mass to provide a fallback when visual evidence is unreliable. CRG learns layer-wise semantic fusion strengths using a reward that balances recognition performance and nuisance exposure. An independently executed reference trajectory on the same episode provides a paired learning signal. Theoretical analysis relates retained evidence and anchor quality to prototype stability and establishes conditions for unbiased on-policy gradient estimation. Experiments on standard, fine-grained, and cross-domain benchmarks show state-of-the-art accuracy, with an average gain of 1.4% over DVLA-RL. The project page is available at https://peacelwh.github.io/TPAMI27-DVLA-RLpp/.

0.69
19

SAFE-KV: Causal Suffix Repair for Behavior-Preserving Non-Prefix KV Cache Reuse in RAG

cs.CL cs.LG ·2026-10-10 ·Xingjian Zhang
publishedcode/data
Abstract · Accepted to Findings of AACL-IJCNLP 2026. 36 pages, 5 figures, 33 tables. Code: https://github.com/XingjianZhang-dev/safe-kv

Retrieval-augmented generation (RAG) reuses the same documents across many requests, making non-prefix KV-cache reuse attractive. Yet a cached document state is not reusable text: it is a context-conditioned causal state whose keys and values depend on the prefix, position, and preceding documents under which it was computed. We formulate non-prefix KV reuse as a behavior-preservation problem with full prefill on the exact prompt as the reference, and introduce SAFE-KV, which repairs a composed cache by re-encoding the causal suffix from a selected start document. On Qwen2.5-7B-Instruct, prefix-conditioned composition preserves only 47.2% of HotpotQA-500 and 37.8% of 2Wiki-500 full-prefill answers; SAFE-KV with risk-guided K2 selection restores 99.8%/98.4% (donor-copy protocol) and 100.0%/98.4% with actual suffix forwarding. Document-only and fixed-window repairs recover far less, because the document states after the repaired region keep histories that omit it. Reuse without repair is exact up to four documents and degrades at sixteen, where suffix repair recovers 98-100%. Faithful repair re-encodes most of the context, so in a single-request L40S study SAFE-KV's operating rule serves every workload by full prefill. SAFE-KV thus establishes the causal-state contract for non-prefix reuse and a validated repair mechanism that meets it.

0.69
20

Do you "see" what I "see"? A Multi-panel Visual Question Answer Dataset for Large Language Model Chart Analysis

cs.DL cs.AIcs.IR ·2026-10-09 ·Daniel J. Evans, J. P. Naiman, JooYoung Seo, J. Stephen Downie
publishedcode/data
Abstract · 16 pages, 4 figures, accepted to The 30th International Conference on Theory and Practice of Digital Libraries, 2026

Digital libraries increasingly host scientific publications whose figures encode primary research findings, yet making this visual content searchable, accessible, and reusable remains an open challenge. Large language and multimodal models (LLMs/LMMs) provide the possibility to automate many figure parsing tasks such as determining data trends or extracting data from plots. In this work, we consider if these models can reliably support task-oriented visual data analysis when presented with synthetic, but realistic, scientific figures that contain multiple charts. Our Visual Question Answering (VQA) process scaffolds our tasks along a four-level description model from basic conceptual answers to interpretive and analytical understanding. Additionally, we outline our process for synthetically generating our single and multi-panel (charts in which there are multiple panels present) plots and testing the data with various LMMs using quantitative analysis. In general, we find LMMs become less accurate with increasing figure complexity as the number of panels, number of plot elements, and difficulty of questions increase. Finally, because LMM accuracy is expected to change rapidly, our analysis also considers cost relative to performance as a metric to inform future user decisions. We provide the detailed question-answer pairs used in this analysis, along with the code used to generate random plots, question-answer pairs and detailed bounding-box metadata for chart elements. The code and data are available at https://github.com/ReadingTimeMachine/LLM_VQA_MultiPanel and https://huggingface.co/datasets/ReadingTimeMachine/visual\_qa\_multipanel

0.68