Clinical Language Model Evaluation

Overview

Clinical language model evaluation asks whether models can answer medical questions, interpret records, support conversations, and assist with decisions without causing harm. Researchers use exam-style benchmarks, clinician-authored rubrics, electronic health records, simulated consultations, and tests that vary language, missing information, patient behavior, or evidence; they also compare expert ratings with automated judges. Evaluation increasingly measures safety, omissions, evidence grounding, uncertainty, consistency, and workflow performance alongside accuracy. A central challenge is that strong scores on static tests or plausible reasoning may not transfer to dynamic, incomplete clinical encounters, while automated judges can disagree with clinicians and require validation.

Latest papers

461 papers56 papers in the last four weeks, up 33% on the four weeks before. 0.4% of all new papers.

Mar 2, 2026

PanCanBench: A Comprehensive Benchmark for Evaluating Large Language Models in Pancreatic Oncology

Large language models (LLMs) have achieved expert-level performance on standardized examinations, yet multiple-choice accuracy poorly reflects real-world clinical utility and safety. As patients and clinicians increasingly use LLMs for guidance on complex conditions such as pancreatic cancer, evaluation must extend beyond general medical knowledge. Existing frameworks, such as HealthBench, rely on simulated queries and lack disease-specific depth. Moreover, high rubric-based scores do not ensure factual correctness, underscoring the need to assess hallucinations. We developed a human-in-the-loop pipeline to create expert rubrics for de-identified patient questions from the Pancreatic Cancer Action Network (PanCAN). The resulting benchmark, PanCanBench, includes 3,130 question-specific criteria across 282 authentic patient questions. We evaluated 22 proprietary and open-source LLMs using an LLM-as-a-judge framework, measuring clinical completeness, factual accuracy, and web-search integration. Models showed substantial variation in rubric-based completeness, with scores ranging from 46.5% to 82.3%. Factual errors were common, with hallucination rates (the percentages of responses containing at least one factual error) ranging from 6.0% for Gemini-2.5 Pro and GPT-4o to 53.8% for Llama-3.1-8B. Importantly, newer reasoning-optimized models did not consistently improve factuality: although o3 achieved the highest rubric score, it produced inaccuracies more frequently than other GPT-family models. Web-search integration did not inherently guarantee better responses. The average score changed from 66.8% to 63.9% for Gemini-2.5 Pro and from 73.8% to 72.8% for GPT-5 when web search was enabled. Synthetic AI-generated rubrics inflated absolute scores by 17.9 points on average while generally maintaining similar relative ranking.
Mar 1, 2026

Prompt Sensitivity and Answer Consistency of Small Open-Source Language Models for Clinical Question Answering in Low-Resource Healthcare

Small open-source language models are gaining attention for healthcare applications in low-resource settings where cloud infrastructure and GPU hardware may be unavailable. However, the reliability of these models under different phrasings of the same clinical question remains poorly understood. We evaluate five open-source models (Gemma 2 2B, Phi-3 Mini 3.8B, Llama 3.2 3B, Mistral 7B, and Meditron-7B, a domain-pretrained model without instruction tuning) across three clinical question answering datasets (MedQA, MedMCQA, and PubMedQA) using five prompt styles: original, formal, simplified, roleplay, and direct. Model behavior is evaluated using consistency scores, accuracy, and instruction-following failure rates. All experiments were conducted locally on consumer CPU hardware without fine-tuning. Consistency and accuracy were largely independent across models. Gemma 2 achieved the highest consistency (0.845-0.888) but the lowest accuracy (33.0-43.5%), while Llama 3.2 showed moderate consistency (0.774-0.807) alongside the highest accuracy (49.0-65.0%). Roleplay prompts consistently reduced accuracy across all models, with Phi-3 Mini dropping 21.5 percentage points on MedQA. Meditron-7B exhibited near-complete instruction-following failure on PubMedQA (99.0% UNKNOWN rate), indicating that domain pretraining alone is insufficient for structured clinical question answering. These findings show that high consistency does not imply correctness: models can be reliably wrong, a dangerous failure mode in clinical AI. Llama 3.2 demonstrated the strongest balance of accuracy and reliability for low-resource deployment. Safe clinical AI requires joint evaluation of consistency, accuracy, and instruction adherence.
Feb 28, 2026

Leveraging Computerized Adaptive Testing for Cost-effective Evaluation of Large Language Models in Medical Benchmarking

The rapid proliferation of large language models (LLMs) in healthcare creates an urgent need for scalable and psychometrically sound evaluation methods. Conventional static benchmarks are costly to administer repeatedly, vulnerable to data contamination, and lack calibrated measurement properties for fine-grained performance tracking. We propose and validate a computerized adaptive testing (CAT) framework grounded in item response theory (IRT) for efficient assessment of standardized medical knowledge in LLMs. The study comprises a two-phase design: a Monte Carlo simulation to identify optimal CAT configurations and an empirical evaluation of 38 LLMs using a human-calibrated medical item bank. Each model completed both the full item bank and an adaptive test that dynamically selected items based on real-time ability estimates and terminated upon reaching a predefined reliability threshold (standard error <= 0.3). Results show that CAT-derived proficiency estimates achieved a near-perfect correlation with full-bank estimates (r = 0.988) while using only 1.3 percent of the items. Evaluation time was reduced from several hours to minutes per model, with substantial reductions in token usage and computational cost, while preserving inter-model performance rankings. This work establishes a psychometric framework for rapid, low-cost benchmarking of foundational medical knowledge in LLMs. The proposed adaptive methodology is intended as a standardized pre-screening and continuous monitoring tool and is not a substitute for real-world clinical validation or safety-oriented prospective studies.
Feb 28, 2026

CURE: A Multimodal Benchmark for Clinical Understanding and Retrieval Evaluation

Multimodal large language models (MLLMs) demonstrate considerable potential in clinical diagnostics, a domain that inherently requires synthesizing complex visual and textual data alongside consulting authoritative medical literature. However, existing benchmarks primarily evaluate MLLMs in end-to-end answering scenarios. This limits the ability to disentangle a model's foundational multimodal reasoning from its proficiency in evidence retrieval and application. We introduce the Clinical Understanding and Retrieval Evaluation (CURE) benchmark. Comprising 500500 multimodal clinical cases mapped to physician-cited reference literature, CURE evaluates reasoning and retrieval under controlled evidence settings to disentangle their respective contributions. We evaluate state-of-the-art MLLMs across distinct evidence-gathering paradigms in both closed-ended and open-ended diagnosis tasks. Evaluations reveal a stark dichotomy: while advanced models demonstrate clinical reasoning proficiency when supplied with physician reference evidence (achieving up to 73.4%73.4\% accuracy on differential diagnosis), their performance substantially declines (as low as 25.4%25.4\%) when reliant on independent retrieval mechanisms. This disparity highlights the dual challenges of effectively integrating multimodal clinical evidence and retrieving precise supporting literature. CURE is publicly available at https://github.com/yanniangu/CURE.
Feb 27, 2026

When Metrics Disagree: Automatic Similarity vs. LLM-as-a-Judge for Clinical Dialogue Evaluation

As Large Language Models (LLMs) are increasingly integrated into healthcare to address complex inquiries, ensuring their reliability remains a critical challenge. Recent studies have highlighted that generic LLMs often struggle in clinical contexts, occasionally producing misleading guidance. To mitigate these risks, this research focuses on the domain-specific adaptation of Llama-2-7B using the Low-Rank Adaptation (LoRA) technique. By injecting trainable low-rank matrices into the Transformer layers, we efficiently adapted the model using authentic patient-physician transcripts while preserving the foundational knowledge of the base model. Our objective was to enhance precision and contextual relevance in responding to medical queries by capturing the specialized nuances of clinical discourse. Due to the resource-intensive nature of large-scale human validation, the model's performance was evaluated through a dual-track framework: Track A utilized traditional lexical similarity metrics (e.g., BLEU, ROUGE), while Track B employed an "LLM-as-a-Judge" paradigm using GPT-4 for semantic assessment. Our results demonstrate that while the LoRA-enhanced model achieved significant improvements across all quantitative lexical dimensions, a profound disagreement surfaced in the GPT-4 evaluation, which marginally favored the baseline model's conversational flow. This metric divergence underscores a pivotal finding: traditional automated scores may not fully reflect clinical utility. Consequently, we propose that while automated metrics and LLM judges serve as valuable developmental proxies, rigorous validation by human medical experts remains an indispensable requirement for the safe deployment of LLMs in healthcare settings.
Feb 26, 2026

MediX-R1: Open Ended Medical Reinforcement Learning

In this work, we introduce MediX-R1, an open-ended Reinforcement Learning (RL) framework for medical vision language models (VLMs) that enables free-form answers beyond multiple-choice formats. MediX-R1 fine-tunes a baseline vision-language backbone with Group Based RL and a composite reward tailored for medical reasoning: an LLM-based accuracy reward that judges semantic correctness with a strict YES/NO decision, a medical embedding-based semantic reward to capture paraphrases and terminology variants, and lightweight format and modality rewards that enforce interpretable reasoning and modality recognition. This multi-signal design provides stable, informative feedback for open-ended outputs where traditional verifiable or MCQ-only rewards fall short. To evaluate performance, we employ a unified evaluation framework for both text-only and multimodal (image+text) tasks that uses a Reference-based LLM-as-judge in place of brittle string-overlap metrics, capturing semantic correctness, clinical accuracy, and contextual alignment. Our results show that open-ended RL with comprehensive reward signals can effectively strengthen medical reasoning in multimodal models. Despite using only ∼51\sim51K instruction examples, MediX-R1 achieves strong performance across standard medical LLM (text-only) and VLM (image + text) benchmarks, outperforming existing open-weight medical VLMs, with large gains particularly on open-ended clinical tasks. These improvements remain consistent across multiple RL algorithms and VLM backbones and are further supported by blinded evaluations with medical experts. Our trained models, curated datasets and source code are available at https://medix.cvmbzuai.com
Feb 26, 2026

Toward Automatic Filling of Case Report Forms: A Case Study on Data from an Italian Emergency Department

Case Report Forms (CRFs) collect data about patients and are at the core of well-established practices to conduct research in clinical settings. With the recent progress of language technologies, there is an increasing interest in automatic CRF-filling from clinical notes, mostly based on the use of Large Language Models (LLMs). However, there is a general scarcity of annotated CRF data, both for training and testing LLMs, which limits the progress on this task. As a step in the direction of providing such data, we present a new dataset of clinical notes from an Italian Emergency Department annotated with respect to a pre-defined CRF containing 134 items to be filled. We provide an analysis of the data, define the CRF-filling task and metric for its evaluation, and report on pilot experiments where we use an open-source state-of-the-art LLM to automatically execute the task. Results of the case-study show that (i) CRF-filling from real clinical notes in Italian can be approached in a zero-shot setting; (ii) LLMs' results are affected by biases (e.g., a cautious behaviour favours "unknown" answers), which need to be corrected.
Feb 26, 2026

ClinDet-Bench: Beyond Abstention, Evaluating Judgment Determinability of LLMs in Clinical Decision-Making

Clinical decisions are often required under incomplete information. Clinical experts must identify whether available information is sufficient for judgment, as both premature conclusion and unnecessary abstention can compromise patient safety. To evaluate this capability of large language models (LLMs), we developed ClinDet-Bench, a benchmark based on clinical scoring systems that decomposes incomplete-information scenarios into determinable and undeterminable conditions. Identifying determinability requires considering all hypotheses about missing information, including unlikely ones, and verifying whether the conclusion holds across them. We find that recent LLMs fail to identify determinability under incomplete information, producing both premature judgments and excessive abstention, despite correctly explaining the underlying scoring knowledge and performing well under complete information. These findings suggest that existing benchmarks are insufficient to evaluate the safety of LLMs in clinical settings. ClinDet-Bench provides a framework for evaluating determinability recognition, leading to appropriate abstention, with potential applicability to medicine and other high-stakes domains, and is publicly available.
Feb 26, 2026

Decomposing Physician Disagreement in HealthBench

We decompose physician disagreement in the HealthBench medical AI evaluation dataset to understand where variance resides and what observable features can explain it. Rubric identity accounts for 15.8% of met/not-met label variance but only 3.6-6.9% of disagreement variance; physician identity accounts for just 2.4%. The dominant 81.8% case-level residual is not reduced by HealthBench's metadata labels (z = -0.22, p = 0.83), normative rubric language (pseudo R^2 = 1.2%), medical specialty (0/300 Tukey pairs significant), surface-feature triage (AUC = 0.58), or embeddings (AUC = 0.485). Disagreement follows an inverted-U with completion quality (AUC = 0.689), confirming physicians agree on clearly good or bad outputs but split on borderline cases. Physician-validated uncertainty categories reveal that reducible uncertainty (missing context, ambiguous phrasing) more than doubles disagreement odds (OR = 2.55, p < 10^(-24)), while irreducible uncertainty (genuine medical ambiguity) has no effect (OR = 1.01, p = 0.90), though even the former explains only ~3% of total variance. The agreement ceiling in medical AI evaluation is thus largely structural, but the reducible/irreducible dissociation suggests that closing information gaps in evaluation scenarios could lower disagreement where inherent clinical ambiguity does not, pointing toward actionable evaluation design improvements.
Feb 25, 2026

Multi-dimensional Assessment and Explainable Feedback for Counselor Responses to Client Resistance in Text-based Counseling with LLMs

Effectively addressing client resistance is a sophisticated clinical skill in psychological counseling, yet practitioners often lack timely and scalable supervisory feedback to refine their approaches. Although current NLP research has examined overall counseling quality and general therapeutic skills, it fails to provide granular evaluations of high-stakes moments where clients exhibit resistance. In this work, we present a comprehensive pipeline for the multi-dimensional evaluation of human counselors' interventions specifically targeting client resistance in text-based therapy. We introduce a theory-driven framework that decomposes counselor responses into four distinct communication mechanisms. Leveraging this framework, we curate and share an expert-annotated dataset of real-world counseling excerpts, pairing counselor-client interactions with professional ratings and explanatory rationales. Using this data, we perform full-parameter instruction tuning on a Llama-3.1-8B-Instruct backbone to model fine-grained evaluative judgments of response quality and generate explanations underlying. Experimental results show that our approach can effectively distinguish the quality of different communication mechanisms (77-81% F1), substantially outperforming GPT-4o and Claude-3.5-Sonnet (45-59% F1). Moreover, the model produces high-quality explanations that closely align with expert references and receive near-ceiling ratings from human experts (2.8-2.9/3.0). A controlled experiment with 43 counselors further confirms that receiving these AI-generated feedback significantly improves counselors' ability to respond effectively to client resistance.
Feb 24, 2026

Small Language Models for Privacy-Preserving Clinical Information Extraction in Low-Resource Languages

Extracting clinical information from medical transcripts in low-resource languages remains a significant challenge in healthcare natural language processing (NLP). This study evaluates a two-step pipeline combining Aya-expanse-8B as a Persian-to-English translation model with five open-source small language models (SLMs) -- Qwen2.5-7B-Instruct, Llama-3.1-8B-Instruct, Llama-3.2-3B-Instruct, Qwen2.5-1.5B-Instruct, and Gemma-3-1B-it -- for binary extraction of 13 clinical features from 1,221 anonymized Persian transcripts collected at a cancer palliative care call center. Using a few-shot prompting strategy without fine-tuning, models were assessed on macro-averaged F1-score, Matthews Correlation Coefficient (MCC), sensitivity, and specificity to account for class imbalance. Qwen2.5-7B-Instruct achieved the highest overall performance (median macro-F1: 0.899; MCC: 0.797), while Gemma-3-1B-it showed the weakest results. Larger models (7B--8B parameters) consistently outperformed smaller counterparts in sensitivity and MCC. A bilingual analysis of Aya-expanse-8B revealed that translating Persian transcripts to English improved sensitivity, reduced missing outputs, and boosted metrics robust to class imbalance, though at the cost of slightly lower specificity and precision. Feature-level results showed reliable extraction of physiological symptoms across most models, whereas psychological complaints, administrative requests, and complex somatic features remained challenging. These findings establish a practical, privacy-preserving blueprint for deploying open-source SLMs in multilingual clinical NLP settings with limited infrastructure and annotation resources, and highlight the importance of jointly optimizing model scale and input language strategy for sensitive healthcare applications.
Feb 23, 2026

Assessing Risks of Large Language Models in Mental Health Support: A Framework for Automated Clinical AI Red Teaming

Large Language Models (LLMs) are increasingly utilized for mental health support; however, current safety benchmarks often fail to detect the complex, longitudinal risks inherent in therapeutic dialogue. We introduce an evaluation framework that pairs AI psychotherapists with simulated patient agents equipped with dynamic cognitive-affective models and assesses therapy session simulations against a comprehensive quality of care and risk ontology. We apply this framework to a high-impact test case, Alcohol Use Disorder, evaluating six AI agents (including ChatGPT, Gemini, and Character AI) against a clinically-validated cohort of 15 patient personas representing diverse clinical phenotypes. Our large-scale simulation (N=369 sessions) reveals critical safety gaps in the use of AI for mental health support. We identify specific iatrogenic risks, including the validation of patient delusions ("AI Psychosis") and failure to de-escalate suicide risk. Finally, we validate an interactive data visualization dashboard with diverse stakeholders, including AI engineers and red teamers, mental health professionals, and policy experts (N=9), demonstrating that this framework effectively enables stakeholders to audit the "black box" of AI psychotherapy. These findings underscore the critical safety risks of AI-provided mental health support and the necessity of simulation-based clinical red teaming before deployment.
Feb 23, 2026

TherapyGym: Evaluating and Aligning Clinical Fidelity and Safety in Therapy Chatbots

Large language models (LLMs) are increasingly used for mental-health support; yet prevailing evaluation methods--fluency metrics, preference tests, and generic dialogue benchmarks--fail to capture the clinically critical dimensions of psychotherapy. We introduce THERAPYGYM, a framework that evaluates and improves therapy chatbots along two clinical pillars: fidelity and safety. Fidelity is measured using the Cognitive Therapy Rating Scale (CTRS), implemented as an automated pipeline that scores adherence to CBT techniques over multi-turn sessions. Safety is assessed using a multi-label annotation scheme, covering therapy-specific risks (e.g., failing to address harm or abuse). To mitigate bias and unreliability in LLM-based judges, we further release THERAPYJUDGEBENCH, a validation set of 116 dialogues with 1,270 expert ratings for auditing and calibration against licensed clinicians. THERAPYGYM also serves as a training harness: CTRS and safety-based rewards drive RL with configurable patient simulations spanning diverse symptom profiles. Models trained in THERAPYGYM improve on expert ratings, with average CTRS rising from 0.10 to 0.60 (and 0.16 to 0.59 under LLM judges). Our work enables scalable development of therapy chatbots that are faithful to evidence-based practice and safer in high-stakes use.
Feb 18, 2026

LiveClin: A Live Clinical Benchmark without Leakage

The reliability of medical LLM evaluation is critically undermined by data contamination and knowledge obsolescence, leading to inflated scores on static benchmarks. To address these challenges, we introduce LiveClin, a live benchmark designed for approximating real-world clinical practice. Built from contemporary, peer-reviewed case reports and updated biannually, LiveClin ensures clinical currency and resists data contamination. Using a verified AI-human workflow involving 239 physicians, we transform authentic patient cases into complex, multimodal evaluation scenarios that span the entire clinical pathway. The benchmark currently comprises 1,407 case reports and 6,605 questions. Our evaluation of 26 models on LiveClin reveals the profound difficulty of these real-world scenarios, with the top-performing model achieving a Case Accuracy of just 35.7%. In benchmarking against human experts, Chief Physicians achieved the highest accuracy, followed closely by Attending Physicians, with both surpassing most models. LiveClin thus provides a continuously evolving, clinically grounded framework to guide the development of medical LLMs towards closing this gap and achieving greater reliability and real-world utility. Our data and code are publicly available at https://github.com/AQ-MedAI/LiveClin.
Feb 17, 2026

Multi-Objective Alignment of Language Models for Personalized Psychotherapy

Mental health disorders affect over 1 billion people worldwide, yet access to care remains limited by workforce shortages and cost constraints. While AI systems show therapeutic promise, current alignment approaches optimize objectives independently, failing to balance patient preferences with clinical safety. We survey 335 individuals with lived mental health experience to collect preference rankings across therapeutic dimensions, then develop a multi-objective alignment framework using direct preference optimization. We train reward models for six criteria -- empathy, safety, active listening, self-motivated change, trust/rapport, and patient autonomy -- and systematically compare multi-objective approaches against single-objective optimization, supervised fine-tuning, and parameter merging. Multi-objective DPO (MODPO) achieves superior balance (77.6% empathy, 62.6% safety) compared to single-objective optimization (93.6% empathy, 47.8% safety), and therapeutic criteria outperform general communication principles by 17.2%. Blinded clinician evaluation confirms MODPO is consistently preferred, with LLM-evaluator agreement comparable to inter-clinician reliability.
Feb 17, 2026

Evidence-Grounded Subspecialty Reasoning: Evaluating a Curated Clinical Intelligence Layer on the 2025 Endocrinology Board-Style Examination

Background: Large language models have demonstrated strong performance on general medical examinations, but subspecialty clinical reasoning remains challenging due to rapidly evolving guidelines and nuanced evidence hierarchies. Methods: We evaluated January Mirror, an evidence-grounded clinical reasoning system, against frontier LLMs (GPT-5, GPT-5.2, Gemini-3-Pro) on a 120-question endocrinology board-style examination. Mirror integrates a curated endocrinology and cardiometabolic evidence corpus with a structured reasoning architecture to generate evidence-linked outputs. Mirror operated under a closed-evidence constraint without external retrieval. Comparator LLMs had real-time web access to guidelines and primary literature. Results: Mirror achieved 87.5% accuracy (105/120; 95% CI: 80.4-92.3%), exceeding a human reference of 62.3% and frontier LLMs including GPT-5.2 (74.6%), GPT-5 (74.0%), and Gemini-3-Pro (69.8%). On the 30 most difficult questions (human accuracy less than 50%), Mirror achieved 76.7% accuracy. Top-2 accuracy was 92.5% for Mirror versus 85.25% for GPT-5.2. Conclusions: Mirror provided evidence traceability: 74.2% of outputs cited at least one guideline-tier source, with 100% citation accuracy on manual verification. Curated evidence with explicit provenance can outperform unconstrained web retrieval for subspecialty clinical reasoning and supports auditability for clinical deployment.
Feb 16, 2026

Fine-Tune, Don't Prompt, Your Language Model to Identify Biased Language in Clinical Notes

Clinical documentation can contain emotionally charged language with stigmatizing or privileging valences. We present a framework for detecting and classifying such language as stigmatizing, privileging, or neutral. We constructed a curated lexicon of biased terms scored for emotional valence. We then used lexicon-based matching to extract text chunks from OB-GYN delivery notes (Mount Sinai Hospital, NY) and MIMIC-IV discharge summaries across multiple specialties. Three clinicians annotated all chunks, enabling characterization of valence patterns across specialties and healthcare systems. We benchmarked multiple classification strategies (zero-shot prompting, in-context learning, and supervised fine-tuning) across encoder-only models (GatorTron) and generative large language models (Llama). Fine-tuning with lexically primed inputs consistently outperformed prompting approaches. GatorTron achieved an F1 score of 0.96 on the OB-GYN test set, outperforming larger generative models while requiring minimal prompt engineering and fewer computational resources. External validation on MIMIC-IV revealed limited cross-domain generalizability (F1 < 0.70, 44% drop). Training on the broader MIMIC-IV dataset improved generalizability when testing on OB-GYN (F1 = 0.71, 11% drop), but at the cost of reduced precision. Our findings demonstrate that fine-tuning outperforms prompting for emotional valence classification and that models must be adapted to specific medical specialties to achieve clinically appropriate performance. The same terms can carry different emotional valences across specialties: words with clinical meaning in one context may be stigmatizing in another. For bias detection, where misclassification risks undermining clinician trust or perpetuating patient harm, specialty-specific fine-tuning is essential to capture these semantic shifts. * Equal contribution.
Feb 16, 2026

Assessing Large Language Models for Medical QA: Zero-Shot and LLM-as-a-Judge Evaluation

Recently, Large Language Models (LLMs) have gained significant traction in medical domain, especially in developing a QA systems to Medical QA systems for enhancing access to healthcare in low-resourced settings. This paper compares five LLMs deployed between April 2024 and August 2025 for medical QA, using the iCliniq dataset, containing 38,000 medical questions and answers of diverse specialties. Our models include Llama-3-8B-Instruct, Llama 3.2 3B, Llama 3.3 70B Instruct, Llama-4-Maverick-17B-128E-Instruct, and GPT-5-mini. We are using a zero-shot evaluation methodology and using BLEU and ROUGE metrics to evaluate performance without specialized fine-tuning. Our results show that larger models like Llama 3.3 70B Instruct outperform smaller models, consistent with observed scaling benefits in clinical tasks. It is notable that, Llama-4-Maverick-17B exhibited more competitive results, thus highlighting evasion efficiency trade-offs relevant for practical deployment. These findings align with advancements in LLM capabilities toward professional-level medical reasoning and reflect the increasing feasibility of LLM-supported QA systems in the real clinical environments. This benchmark aims to serve as a standardized setting for future study to minimize model size, computational resources and to maximize clinical utility in medical NLP applications.
Feb 2, 2026

LingLanMedBench: A Unified Benchmark for Evaluating Large Language Models in Traditional Chinese Medicine

Reliable evaluation of large language models in traditional Chinese medicine remains limited because existing benchmarks mainly focus on examination-style question answering or isolated generative tasks. We developed LingLanMedBench, a unified benchmark comprising 25,620 expert-verified items across 13 subtasks and five domains, covering knowledge, language understanding, information extraction, and diagnostic-therapeutic decision-making. LingLanMedBench standardizes task formats and evaluation metrics and further incorporates synonym-tolerant clinical-label matching, empirically derived hard subsets, and decision-recognition reformulations of open-ended clinical tasks. Fourteen open-source and proprietary models were evaluated under a standardized zero-shot protocol. Performance varied substantially across domains, model families, task formulations, and difficulty levels. The leading overall score decreased from 51.1 on the full benchmark to 31.9 on the hard subsets. Prescription performance also differed markedly between open-ended generation, with F1 scores of 32.7 and 20.3, and decision recognition, with accuracies of 90.6 and 56.8 on the full and hard subsets, respectively. These findings reveal a substantial gap between knowledge-oriented performance and clinically faithful reasoning. LingLanMedBench provides a standardized and reproducible framework for characterizing current capabilities, identifying systematic limitations, and tracking meaningful progress in large language models for traditional Chinese medicine.
Jan 26, 2026

Health-SCORE: Improving Open-Ended Health LLMs Using Scalable Rubric-Based Rewards

Reinforcement learning is a powerful approach for improving language models, but applying it to open-ended domains such as healthcare remains challenging because high-quality responses are often multidimensional, context-dependent, and difficult to capture with a single reward signal. Rubrics offer a natural form of supervision by providing a structured representation of the criteria that define a strong response. Existing rubric-based approaches, however, often rely on instance-specific criteria, creating two important limitations: (1) such rubrics are typically unavailable for unseen examples at test time; and (2) constructing new rubrics for every instance incurs substantial reward design costs. We introduce Health-SCORE, a scalable framework that captures recurring principles of response quality in the form of reusable meta-rubrics. Rather than relying on manually constructed instance-level criteria, Health-SCORE captures generalizable principles that can be applied across examples, tasks, and healthcare datasets. These meta-rubrics support two forms of model improvement: they can serve as structured, safety-aware reward signals for reinforcement learning during post-training, and they can be incorporated directly into prompts to guide model behavior at inference time. Across several open-ended healthcare benchmarks, models trained with Health-SCORE achieve performance competitive with models trained using human-authored instance-level rewards, while substantially reducing reward-design costs and improving generalizability.
Jan 25, 2026

Health-ORSC-Bench: A Benchmark for Measuring Over-Refusal and Safety Completion in Health Context

Safety alignment in Large Language Models is critical for healthcare; however, reliance on binary refusal boundaries often results in over-refusal of benign queries or unsafe compliance with harmful ones. While existing benchmarks measure these extremes, they fail to evaluate Safe Completion: the model's ability to maximise helpfulness on dual-use or borderline queries by providing safe, high-level guidance without crossing into actionable harm. We introduce Health-ORSC-Bench, the first large-scale benchmark designed to systematically measure Over-Refusal and Safe Completion quality in healthcare. Comprising 31,920 benign boundary prompts across seven health categories (e.g., self-harm, medical misinformation), our framework uses an automated pipeline with human validation to test models at varying levels of intent ambiguity. We evaluate 30 state-of-the-art LLMs, including GPT-5 and Claude-4, revealing a significant tension: safety-optimised models frequently refuse up to 80% of "Hard" benign prompts, while domain-specific models often sacrifice safety for utility. Our findings demonstrate that model family and size significantly influence calibration: larger frontier models (e.g., GPT-5, Llama-4) exhibit "safety-pessimism" and higher over-refusal than smaller or MoE-based counterparts (e.g., Qwen-3-Next), highlighting that current LLMs struggle to balance refusal and compliance. Health-ORSC-Bench provides a rigorous standard for calibrating the next generation of medical AI assistants toward nuanced, safe, and helpful completions. Furthermore, our benchmark facilitates reproducible evaluation, encourages safety calibration, and supports development of clinically reliable, context-aware, human-aligned medical AI systems. Our code and data are available at: https://github.com/ZhihaoZhang97/Health-ORSC-Bench. Warning: Some contents may include toxic or undesired contents.
Jan 23, 2026

SycoEval-EM: Sycophancy Evaluation of Large Language Models in Simulated Clinical Encounters for Emergency Care

Large language models (LLMs) deployed in clinical decision support may acquiesce to patient requests for care that conflicts with evidence-based guidelines. We developed SycoEval-EM, a multi-agent simulation framework to evaluate LLM robustness to adversarial patient persuasion in emergency medicine. Across 19 contemporary LLMs and 1,425 simulated clinical encounters spanning three Choosing Wisely scenarios, acquiescence rates ranged from 0% to 100%, revealing a bimodal distribution. Seven models maintained near-perfect guideline adherence, while six acquiesced in the majority of encounters. Vulnerability varied substantially across clinical scenarios. Acquiescence was highest for CT imaging requests, intermediate for antibiotic prescriptions for sinusitis, and lowest for opioid prescriptions for acute back pain. Model scale, recency, and performance on static medical benchmarks did not consistently predict robustness. All five persuasion tactics produced similar acquiescence rates, with no statistically significant differences after correction for multiple comparisons, suggesting a generalized susceptibility rather than tactic-specific weaknesses. LLM-as-judge evaluation was validated against two independent physician raters across 95 matched conversations and demonstrated near-perfect agreement for the primary outcome of acquiescence (Cohens kappa = 0.957). These findings indicate that static medical benchmarks are insufficient to predict safety performance under sustained social pressure and support incorporating multi-turn adversarial testing into clinical AI evaluation. Notably, two models achieved perfect guideline adherence across all encounters, demonstrating that robustness to patient pressure is attainable without sacrificing effective clinical communication.
Jan 22, 2026

Towards Reliable Medical LLMs: Benchmarking and Enhancing Confidence Estimation of Large Language Models in Medical Consultation

Large-scale language models (LLMs) often offer clinical judgments based on incomplete information, increasing the risk of misdiagnosis. Existing studies have primarily evaluated confidence in single-turn, static settings, overlooking the coupling between confidence and correctness as clinical evidence accumulates during real consultations, which limits their support for reliable decision-making. We propose the first benchmark for assessing confidence in multi-turn interaction during realistic medical consultations. Our benchmark unifies three types of medical data for open-ended diagnostic generation and introduces an information sufficiency gradient to characterize the confidence-correctness dynamics as evidence increases. We implement and compare 27 representative methods on this benchmark; two key insights emerge: (1) medical data amplifies the inherent limitations of token-level and consistency-level confidence methods, and (2) medical reasoning must be evaluated for both diagnostic accuracy and information completeness. Based on these insights, we present MedConf, an evidence-grounded linguistic self-assessment framework that constructs symptom profiles via retrieval-augmented generation, aligns patient information with supporting, missing, and contradictory relations, and aggregates them into an interpretable confidence estimate through weighted integration. Across two LLMs and three medical datasets, MedConf consistently outperforms state-of-the-art methods on both AUROC and Pearson correlation coefficient metrics, maintaining stable performance under conditions of information insufficiency and multimorbidity. These results demonstrate that information adequacy is a key determinant of credible medical confidence modeling, providing a new pathway toward building more reliable and interpretable large medical models.
Jan 6, 2026

MAPLE: Medical Aspect-Based Summarization with Phrase-Level Evidence

Trustworthy clinical summarization requires every claim to be traceable to its evidence, yet existing attribution often resolves only to the sentence or document, leaving clinicians to scan surrounding text for the few words that matter. We argue that the unit of attribution should match the unit of verification: the precise phrase the reader's eye must land on. We present MAPLE (Medical Aspect-Based Summarization with Phrase-Level Evidence), a human-annotated benchmark that grounds each summarized claim in both cited sentences and contributory phrases within them. Spanning 152 randomized controlled trial (RCT) abstracts and 16 clinically motivated aspects, MAPLE comprises 1,799 aspect-based summaries with two-level evidence. We further introduce a decoupled evaluation framework that separately scores content, traceability, and locatability, together with a proxy for the amount of source text a clinician must inspect to verify a claim. Benchmarking eleven LLMs shows that sentence-level citation is consistently strong (C-F1 up to 90.9%), while phrase-level grounding remains less stable and the most discriminative axis across models (P-F1 66.1-84.5%). These results suggest that the key challenge is not only producing accurate summaries, but localizing their supporting evidence precisely enough for efficient clinical verification. Data and code are available at https://github.com/chubohao/maple.
Dec 25, 2025

Context-Aware Classification and Grading of Sensitive Information in Online Conversational Health Data

Online medical consultations contain sensitive health information whose privacy implications depend not only on the entities mentioned but also on how those entities are described in context. Existing classification and grading approaches often map health-information entities directly to predefined sensitivity levels, potentially overlooking whether a condition is confirmed, suspected, negated, hypothetical, or merely planned for investigation. In this study, we formulate sensitive-information grading in online medical dialogues as a context-aware evaluation task. We develop a standard-informed operational framework that incorporates assertion status, experiencer, test-result status, and information granularity. We further design a naturalistic evaluation setting together with contrastive cases that minimally alter negation, uncertainty, experiencer, or granularity, and compare large language models under mention-only and full-context conditions. The study aims to quantify the contribution of contextual information to sensitivity grading and to characterize safety-critical over- and under-grading errors. Our framework provides a reproducible basis for evaluating whether LLMs can distinguish sensitive entity mentions from contextually established sensitive disclosures.
Dec 1, 2025

First, do NOHARM: a medical safety benchmark and randomized study of physician and AI teaming on clinical consultations

Large language models (LLMs) and medical AI tools are routinely used by physicians and patients for medical advice, yet their clinical safety profiles remain poorly characterized. We present NOHARM (Numerous Options Harm Assessment for Risk in Medicine), a 1,100-task benchmark of primary care-to-specialist consultation cases to measure the frequency and severity of potentially harmful errors from LLM-generated medical consultation recommendations. NOHARM covers 10 specialties, with 12,747 expert annotations for 4,249 clinical management options. Across 20 notable LLMs and 4 widely used retrieval-augmented generation (RAG) clinical AI tools, direct application of recommendations carried potential for severe harm in up to 24.6% of cases, with errors of omission accounting for more than 80% of severe errors. Harm potential was not uniform across systems, with clinical AI tools outperforming generalist LLMs, and multi-agent AI teaming further improving performance in generalist models. In a randomized study of 101 U.S.-licensed generalist physicians, AI assistance improved physician performance compared to conventional resources. However, AI-assisted physicians frequently omitted valuable AI-generated recommendations and still scored lower than many AI systems alone. Had those recommendations been incorporated, combined human-AI responses would have outperformed both the human and AI system as used, suggesting complementary strengths and unrealized potential in human-AI teaming. Collectively, these results show that despite strong performance on medical knowledge benchmarks, widely used AI tools can produce medical consultation advice with the potential for severe harm, and highlight the need for explicit measurement of clinical safety. The benchmark and leaderboard are publicly available to support ongoing evaluation and improvement of AI systems used for clinical care.
Nov 20, 2025

Algorithmic Reflectivity in Large Language Models: Methodological Study of Linguistic Structure in Mentalization

Mentalization is a core human capacity involving the interpretation of one's own and others' behavior in terms of underlying mental states. Large language models (LLMs) can generate linguistically reflective discourse, raising the question of whether the formal linguistic structure of mentalization can be reproduced without experiential and affective processes. This study assessed whether an LLM can generate mentalization profiles structurally coherent with the dimensions of mentalization-based treatment (MBT) and recognized as formally mentalizing by expert clinicians. Fifty human-model dialogues were generated with GPT-4.1 under standardized conditions; at the end of each, the model produced a narrative mentalization profile. Five MBT-trained psychiatrists independently rated each profile on 4 canonical MBT axes and 2 higher-order dimensions (regulatory functions, synthetic profile) for evaluative and argumentative coherence. Interrater reliability was assessed with ICC(3,k). Mean scores ranged from 3.63 to 3.98, indicating moderate-to-high structural coherence; ICC(3,k) in the evaluative domain ranged from 0.60 to 0.84. Evaluative and argumentative coherence were strongly associated (Pearson r 0.826-0.967). A negative control with mismatched dialogue-profile pairs yielded minimal ratings. Clinicians described outputs as linguistically organized and clinically interpretable, but with limited affective variability and contextual integration. The findings suggest that LLMs can reproduce the formal linguistic structure of reflective discourse - a phenomenon we term algorithmic reflectivity - which appears more consistent with structural language regularities than with experiential or intentional processes. LLMs may serve as research and training tools for studying the linguistic aspects of mentalization.
Nov 20, 2025

Knowledge and Reasoning Augmented Learning for Language Model Assisted Antimicrobial Decision Support

Background: Antimicrobial decision support requires both clinical knowledge and patient-specific reasoning. We developed Knowledge and Reasoning Augmented Learning (KRAL) to improve language model performance on antimicrobial decision-making tasks. Methods: In this retrospective technical study, a 27-billion-parameter student model underwent guideline-grounded supervised fine-tuning followed by Multi-turn Agentic GRPO using clinical reasoning trajectories. Six configurations were evaluated on 123 clinician-validated clinical antimicrobial treatment cases from Peking Union Medical College Hospital (PUMCH) and 100 JAMA Clinical Challenge cases. A language model judge scored responses, and paired tests compared configurations. Component ablations assessed benchmark performance and training stability. Results: On the PUMCH benchmark, KRAL achieved the highest composite score (0.895 on a 0-1 scale), outperforming all five controls (all P<0.001). On JAMA, KRAL scored 0.600 and exceeded the base model with and without retrieval (P = 0.0048 and 0.0499, respectively). Reinforcement learning ablation experiments demonstrate that the reward smoothing can significantly improve the performance and training stability of vanilla GRPO. Conclusions: KRAL improved performance on a single-center clinical antimicrobial benchmark and publicly available JAMA clinical challenge cases by combining guideline-grounded learning with Multi-turn Agentic GRPO. Further external clinical validation is needed to evaluate the practical utility and effectiveness of KRAL.
Nov 1, 2025

MedRECT: A Bilingual Medical Reasoning Benchmark for Error Correction in Clinical Texts

Large language models (LLMs) show promise in medical applications, but their ability to detect and correct errors in clinical texts remains under-evaluated, particularly beyond English. We introduce MedRECT, a bilingual benchmark for Japanese and English that formulates medical error handling as three subtasks: error detection, error sentence extraction, and error correction. MedRECT-ja contains 663 samples derived from the Japanese Medical Licensing Examinations, while the separately sourced MedRECT-en contains 458 samples curated from MEDEC. We evaluate 11 LLMs across 17 configurations that cover proprietary and open-weight models, medical-domain specialization, and multiple reasoning settings. Qwen3-32B scores higher in its thinking mode than in its non-thinking mode on error detection F1 and sentence extraction accuracy in both subsets, with sentence extraction accuracy higher by 24.5 percentage points on MedRECT-ja and 10.3 on MedRECT-en. Several leading general-purpose reasoning models outperform all three evaluated medical-domain models on these two subtasks. Most models have lower point estimates on the Japanese subset, although absolute scores are not directly comparable because the subsets differ in source material and error distributions. LoRA fine-tuning yields higher sentence extraction accuracy and higher point estimates on all three reference-based correction similarity metrics in both languages. MedRECT provides an open, reusable evaluation resource for studying medical error correction and reasoning across Japanese and English. Our dataset and code are available at https://github.com/pfnet-research/medrect.
Oct 20, 2025

Zoom In Disparities in Healthcare LLM Q&A

Equitable access to reliable health information is vital when integrating AI into healthcare. Yet, information quality varies across languages, raising concerns about the reliability and consistency of multilingual Large Language Models (LLMs). We systematically examine cross-lingual disparities in pre-training source and factuality alignment in LLM answers for multilingual healthcare Q&A across English, German, Turkish, Chinese (Mandarin), and Italian. We (i) constructed Multilingual Wiki Health Care (MultiWikiHealthCare), a multilingual dataset from Wikipedia; (ii) analyzed cross-lingual healthcare coverage; (iii) assessed LLM response alignment with these references; and (iv) conducted a case study on factual alignment through the use of contextual information and Retrieval-Augmented Generation (RAG). Our findings reveal substantial cross-lingual disparities in both Wikipedia coverage and LLM factual alignment. Across LLMs, responses align more with English Wikipedia, even when the prompts are non-English. Providing contextual excerpts from non-English Wikipedia at inference time effectively shifts factual alignment toward culturally relevant knowledge. These results highlight practical pathways for building more equitable, multilingual AI systems for healthcare.