CS Paper Radar

Language-model ratings of depression reflect the rater more than the patient

Baihan Lin

cs.CL cs.AIq-bio.NC ·submitted 2026-10-06 15:08 UTC · v1

Abstract

Depression has no diagnostic blood test. Language models promise tireless, consistent assessment, but can accurate raters disagree about individuals? We pre-registered 880 language-model raters, crossing 11 open models with prompting and scoring choices, and applied them to 189 interviews against the eight-item Patient Health Questionnaire. Model choice explained 30.0% of summed-symptom score variance, stable participant differences 10.5%. Two randomly drawn raters with area under the receiver operating characteristic curve (AUC) >= 0.70 disagreed on screening decisions for 40% of participants, on average. Average over-rating governed how many were flagged, yet equal-capacity raters chose differently for about one participant in five. A locked analysis of 86 new interviews reproduced the main pre-registered findings. Exploratory recalibration with 40 labelled participants raised accuracy from about 60% to 75% and halved disagreement, leaving one participant in five decided differently. Calibration repaired much of the rater dependence without securing agreement about individuals.

Closest papers

Nearest neighbours by meaning across the whole index.

EviDial: When Correct Depression Predictions Lack Evidence

cs.CL ·2026-10-20 ·Junjie Xu, Jin Zhu, Yuzhe Yang, Luhan Wang, Xingjiao Wu +4 more
benchmarkhot direction similarity 0.64
Abstract

Depression dialogue benchmarks commonly assess models by comparing predicted symptom labels with clinical scale scores. However, a prediction may agree with a benchmark label even when the dialogue lacks enough evidence to justify it. We introduce EviDial, a benchmark that distinguishes label agreement from evidence-grounded assessment. EviDial comprises 382 depression dialogues from three sources and 2,744 participant-symptom cells, each classified as having absent, partial, conflicting, or sufficient evidence. Because evidence sufficiency lacks a directly observable gold standard, we derive calibrated evidence references from multiple judgments using generalizability theory, many-facet Rasch measurement, and random-effects latent class analysis. Our audit reveals that 82.8% of participant-symptom cells lack sufficient textual evidence for the corresponding symptom assessment, with substantial variation across sources. Across model families and prompting strategies, 69.2%-96.9% of predictions judged correct by benchmark labels are nonetheless unsupported by sufficient dialogue evidence. Further analyses show that strong agreement among LLM evaluators need not reflect independent evidence judgments, underscoring the need to calibrate automated reference panels. These findings demonstrate that label-based evaluation can overestimate evidence-grounded performance by conflating agreement with clinical labels and support in the observed dialogue. EviDial provides a measurement layer for separating these constructs and auditing LLM-based depression assessment more faithfully.

0.28

K-Bench: a clinically calibrated benchmark for evaluating large language models in high-risk mental health conversations

cs.CL cs.AIcs.LG ·2026-09-14 ·Laura M. Vowels, Matthew J. Vowels, Shivali Sharma, Apoorv Jha, Rehnuma Choudhury +7 more
benchmark similarity 0.63
Abstract

People increasingly use large language models (LLMs) for mental health support, yet their safety in evolving, high-risk conversations remains poorly characterised. We developed K-Bench, a clinician-calibrated, protected benchmark evaluating 125 model configurations representing 33 base models from 14 providers across a fixed cohort of 200 multi-turn vignettes involving suicide, self-harm, domestic violence, substance misuse, and no-risk presentations. Synthetic patient conversations showed substantial distributional overlap with real human-AI conversations. A frozen GPT-4o judge achieved 94.2% exact agreement with clinician consensus across 6,751 eligible item comparisons from 151 clinician-rated transcripts. Leading models combined strong supportive conversation with combined-risk scores above 95, whereas risk exploration exposed substantial variation among lower-performing configurations. Therapeutic prompting produced configuration-specific gains concentrated among weaker models, while elevated reasoning produced no average improvement. K-Bench combines broader clinical coverage and configuration-scale comparison with a continuously updated public leaderboard whose operational test materials are protected from direct optimisation. The leaderboard is available at www.k-bench.ai.

0.32

A Stratified Multi-Rater Evaluation of LLM-Based Virtual Standardized Patients with a Deployed Data-Generation Platform

cs.HC ·2026-10-21 ·Yongjin Yi, Jinyong Park, Sejoong Kim
top venuecode/data similarity 0.62
Abstract · Accepted at NeurIPS 2026, Track on Evaluations and Datasets. 39 pages (10 pages of main text), 3 figures and 3 tables in the main text, 4 fi…

The Objective Structured Clinical Examination (OSCE) is widely used to assess clinical communication skills in medical education, requiring trainees to take histories and reason clinically with standardized patients (SPs). Large language model (LLM)-based virtual standardized patients (VSPs), typically built by persona-injected prompting, offer a practical alternative. However, existing evaluations rely on a single backbone LLM, small human rater panels, or LLM-as-a-judge scoring without an utterance-aligned human-expert reference, and so cannot identify the specific shortcomings of individual models. We evaluated four LLMs (GPT-4o, Claude 3.5 Sonnet, Llama-3.1 70B, and HyperCLOVA X), spanning commercial, open-source and Korean-developed models, against a human-expert reference across three Korean OSCE cases (hematuria, fever in pregnancy, and chronic cough). In a double-blind, fully crossed design, 17 senior medical-student raters scored every model utterance on three sentence-level and six encounter-level Likert metrics, yielding 7,480 data points. We analyzed the scores using a stratified linear mixed-effects model based on an 11-category history-taking and 3-level linguistic-form taxonomy. At the encounter level, three of the four LLMs fell 0.08 to 0.47 Likert points below the human-expert reference on Comprehensibility, Consistency and Hallucination-freeness. The difference was not statistically significant for Hallucination-freeness in Claude 3.5 Sonnet and GPT-4o, Consistency in GPT-4o, and Engagement in HyperCLOVA X. Llama-3.1 70B fell significantly below the reference on all six metrics. We also introduce a web-based OSCE platform that supports virtual practice across diverse clinical cases and continuously collects student-VSP dialogues under research and privacy consent.

0.55

Augmenting Interviewer Judgments of Patient Experience with Automatic Language Analysis

cs.HC cs.CL ·2026-08-31 ·Aowen Shi, Michal Balazia, Danilo Postin, René Hurlemann, Jan Alexandersson +2 more
published similarity 0.60
Abstract · Accepted to ACII 2026 as an oral presentation

Understanding how psychiatric patients subjectively experienced a clinical conversation is important for feedback and alliance-related process monitoring. While interviewers form post-session judgments about patient experience, these judgments do not always match patients' self-reports. Automatic approaches for predicting perceived interaction quality from conversation have been proposed, but it remains unclear whether such approaches can complement human judgment rather than simply replicate it. To address this gap, we evaluate a clinician-support framework in which post-session interviewer ratings are combined with automatic language-based predictions to estimate patient-reported interaction quality in free clinical interviews. We assess this integration across multiple standard model types, including Ridge, SVR, MLP, GRU, and BiLSTM, all trained on sentence embeddings extracted from dyadic transcripts of 107 free conversations between psychiatric patients and interviewers. Our results show that combining interviewer judgments with model predictions through simple averaging yields the strongest overall performance. The interviewer-only baseline reached a Pearson correlation of 0.365. Among fully automatic models, Ridge achieved the strongest Pearson correlation (r = 0.286), while BiLSTM achieved r = 0.270. The strongest result was obtained by BiLSTM interviewer integration (r = 0.403). Our findings suggest that automatic language analysis and interviewer judgment capture complementary aspects of patient experience and that their combination provides a more accurate approximation of the patient's own report than either source alone.

0.34

Can Large Language Model-Generated Responses Support Assessment Development? A Human-Calibrated Rasch Benchmark

cs.CY stat.AP ·2026-09-10 ·Eunjeong Song, Sehee Hong
benchmark similarity 0.60
Abstract · 49 pages, 3 figures, 12 tables. Data, materials, and code: https://doi.org/10.7910/DVN/S7QUCA

Large language models (LLMs) are proposed as synthetic respondents for pilot testing, but their usefulness depends on whether they supply the evidence assessment development requires. We calibrated rating scale models on 14 digital-use skill items from 6,245 adults and used the human item parameters to evaluate responses generated for 1,300 demographically matched personas. LLM responses had high internal consistency ($α\approx .94$) but used the lowest category in 0.2-0.3% of responses versus 13.7-21.5% for humans, and no persona chose it on every item. These gaps changed the response-category judgment on both subscales and the PC targeting judgment; on the human-calibrated scales, 12 of 14 items had infit below 0.70, indicating responses more predictable than the Rasch model expects. Preregistered changes to the prompt, category order, and persona information did not restore the human lower range. High internal consistency is insufficient evidence that LLM responses can replace human pilot data.

0.31

Being judged through the record: A Rosenhan-inspired test of psychiatric label sensitivity in large language models

cs.CY ·2026-10-26 ·Shaked Rosenblum, Teddy Lazebnik, Ziv Ben-Zion
similarity 0.60
Abstract

The integration of large language models (LLMs) into clinical decision-making raises unresolved questions regarding diagnostic bias. We investigate whether a documented psychiatric history alters LLM recommendations when the current clinical evidence is identical. Using a counterfactual experiment inspired by Rosenhan's 1973 field study, we factorially varied diagnostic-history condition (no prior diagnosis stated vs. prior schizophrenia diagnosis) and current-evidence condition (voices ceased vs. voices persisted) within a clinical vignette. Across five LLMs (GPT-5.6 Sol, Claude Sonnet 5, Gemini 3.6 Flash, Grok 4.6, and Mistral Large 2512), we generated 20,000 model responses across independent queries. When evidence supported discharge, adding a prior schizophrenia diagnosis increased recommendations for continued hospitalization nearly fivefold, from 4.7% to 22.3% (risk difference, 17.5 percentage points; 95% CI, 16.2-18.8). This shift occurred across all five models (risk-difference range, 10.4-34.4 percentage points). In the same condition, reference concordance decreased from 95.3% to 77.7%, while stated decision confidence increased. Across all model-by-evidence contrasts, confidence increases corresponded to Cohen's d values of 0.35-1.09. Signal-detection analyses showed that the prior diagnosis shifted the decision criterion toward hospitalization across all five models and reduced discrimination between the two current-evidence conditions in three of them. When current evidence supported continued hospitalization, the label increased the recommendation for hospitalization from 92.6% to 97.0%.

0.17

Whose Assessment of Distress? Community Perspectives and LLM Alignment on Well-Being Posts

cs.CL ·2026-08-29 ·Andrew Aquilina, Xiang Lorraine Li, Yu-Ru Li
similarity 0.59
Abstract

Judgments about psychological distress are socially situated: what counts as concerning hinges on community norms around emotional expression, vulnerability, and help-seeking. Yet large language models (LLMs) used for distress detection are typically aligned to a single, undifferentiated standard. How well do these models capture the perspectives of the communities whose language they assess? We address this question through a perspectivist annotation study in which 321 participants provided 9,587 judgments on 1,198 Reddit posts spanning six identity-based communities, yielding community-specific labels. Raters in the contextualized in-group condition show a modest tendency to agree more with their community than uncontextualized out-group raters (OR = 1.18), an effect varying significantly across communities. We then evaluate nine open-weight LLM configurations and four frontier configurations against these labels. Open-weight LLMs systematically over-estimate distress: when communities perceive none-to-mild distress, these models achieve only 31-44% accuracy, predominantly producing false positives. GPT-5 and Gemini 2.5 Pro show the same none-to-mild inflation even when their full-sample over/under rates are mixed, while Claude Opus 4 is more conservative. This pattern does not simply mirror an outsider reading position: uncontextualized out-group human aggregates were nearly symmetric, with 18% over-estimation versus 19% under-estimation. Instead, the models that inflate none-to-mild cases exhibit a distress prior that exceeds both contextualized in-group and uncontextualized out-group human judgments. These findings have implications for equitable AI deployment in mental health contexts, where miscalibrated distress detection may unevenly affect the communities being assessed.

0.24

ROUNDS: Source-Framed Diagnostic Revision after Multi-Turn Clinical Evidence

cs.CL cs.AI ·2026-11-14 ·Jinrui Fang, Runhan Chen, Yining Wu, Jiawei Xu, Xu Yang +4 more
top venuebenchmark similarity 0.59
Abstract · Accepted at Machine Learning for Health (ML4H) 2026, Proceedings Track; to appear in Proceedings of Machine Learning Research (PMLR). 17 pag…

Large language models (LLMs) achieve high scores on medical benchmarks, but does this performance translate into reliable diagnosis as a conversation unfolds? In diagnostic conversations, users can provide clinical findings across multiple turns, ask for a diagnosis without supplying answer options, and introduce opinions from others. Models must therefore form diagnoses from the available evidence and decide how to respond to subsequent advice. We introduce ROUNDS, a framework that evaluates these abilities after a fixed sequence of clinical evidence delivered over multiple turns. ROUNDS compares diagnosis with and without answer options, presents the same incorrect diagnosis under clinician and lay framing, and varies support for competing diagnoses in written five-clinician panels. Across seven models and 1,035 cases from five datasets, average accuracy was 76.9% with answer options and 57.0% without them. Clinician-framed misleading advice produced lower final accuracy than lay-framed advice in every model, with differences of 4.6 to 38.2 percentage points. Initially correct diagnoses more often became incorrect answers than abstentions, and many adopted the suggested disease. Conversely, accuracy increased as more panel members supported the correct diagnosis, including when the majority opinion remained unchanged. These findings reveal a gap between medical benchmark performance and reliable diagnosis in conversation: models respond to external opinions, but remain vulnerable to abandoning correct diagnoses without new clinical evidence.

0.34