arXiv AI July 21, 2026 By Koyar Afrasyab

Judge-dependent safety gains and model-specific helpfulness costs of evidence-sufficiency prompting in clinical LLMs

arXiv:2607. 18086v1 Announce Type: new Abstract: Background: LLM judges increasingly score whether clinical language models give overconfident answers under incomplete evidence, yet whether a measured "safety gain" reflects real behavior change or the judge's calibration is unresolved.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv AI

Jul 22

Evaluating medical AI under missing information: same-provider judges and human raters change apparent safety

arXiv:2607. 18828v1 Announce Type: new Abstract: Readiness stress-testing of medical AI has focused on closed-ended and multimodal benchmarks.

By Koyar Afrasyab

llms multimodal benchmarks safety

arXiv Machine Learning

Jun 5

IatroBench: Pre-Registered Evidence of Iatrogenic Harm from AI Safety Measures

arXiv:2604. 07709v4 Announce Type: replace-cross Abstract: A heavily safety-trained model will hand a physician the full, patient-followable benzodiazepine taper and refuse it to the patient who needs it, over identical clinical facts; the knowledge is present either way.

By David Gringras

llms benchmarks safety

arXiv AI

Aug 26

A Judge Should Know What Changed:Construct Validity for LLM-as-a-Judge Evaluation

The paper introduces a two‑dimensional construct validity framework for evaluating large language models (LLMs) as judges, defining invariance (S) and sensitivity (R) to construct‑preserving and construct‑changing edits. Experiments across seven judges and four domains reveal high invariance (average S = 0.945) but low sensitivity (average R = 0.319), with sensitivity varying by edit type. Audits of public label sets show that surface‑only predictors can reproduce a substantial portion of labels, underscoring that high agreement does not guarantee construct validity.

By Jianlin Chen, Wenhui Chen, Ziyao Lin, Chi Man Vong

llms

arXiv AI

Aug 3

Reasoning in Real World Clinical Care: Why Large Language Models Are Not Yet Safe for Autonomous Clinical Decision Support

arXiv:2607. 28677v1 Announce Type: new Abstract: LLM now pass medical licensing examinations and, in curated cases, can rival physicians at diagnostic reasoning.

By Shayndhan Sivanathan, Shravan Nageswaran, Mehdi Zadem, Ryaan Sultan, Nicolas von Mallinckrodt, Max Solovyev, Alexey Matyushkin, Sumon Sadhu, Gabriele C DeLuca, Sanjeeva Jeyaretna, James Hillis, Manoj Ramachandran, Prakash Jayakumar

llms agents safety

arXiv AI

Sep 1

A collective capability boundary in frontier large language models on guideline-conformant and case-specific oncology decision-making

arXiv:2608.28592v1 Announce Type: new Abstract: Large language models (LLMs) achieve high scores on medical knowledge examinations, yet real-world oncology is not a knowledge test--it is a sequence o...

By Zhang Sheng, Jinming Li, Wangyang Chen, Zhiwei Bao, Yu YoSean Wang

llms benchmarks

arXiv AI

Aug 5

Preferred, Not Safer: Pairwise Preference Is a Poor Proxy for Clinical Safety

arXiv:2608. 02617v1 Announce Type: cross Abstract: We evaluate whether clinician pairwise preferences provide a reliable signal of clinical safety in large language model (LLM) evaluation using expert feedback from MOOVE (Massive Open Online Validation and Evaluation), a clinician-led platform collecting blinded pairwise preferences alongside multi-criterion rubric ratings.

By Fay Elhassan, David Sasu, Alexandra Kulinkina, Lars Henning Klein, Mary-Anne Hartley

llms safety

One email a morning, machine-written

Related stories

Evaluating medical AI under missing information: same-provider judges and human raters change apparent safety

IatroBench: Pre-Registered Evidence of Iatrogenic Harm from AI Safety Measures

A Judge Should Know What Changed:Construct Validity for LLM-as-a-Judge Evaluation

Reasoning in Real World Clinical Care: Why Large Language Models Are Not Yet Safe for Autonomous Clinical Decision Support

A collective capability boundary in frontier large language models on guideline-conformant and case-specific oncology decision-making

Preferred, Not Safer: Pairwise Preference Is a Poor Proxy for Clinical Safety