Pith. sign in

REVIEW 8 cited by

CARES: Comprehensive Evaluation of Safety and Adversarial Robustness in Medical LLMs

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2505.11413 v1 pith:QKRXK6LM submitted 2025-05-16 cs.CL

CARES: Comprehensive Evaluation of Safety and Adversarial Robustness in Medical LLMs

classification cs.CL
keywords safetyadversarialcaresmedicalevaluationllmspromptsassess
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Large language models (LLMs) are increasingly deployed in medical contexts, raising critical concerns about safety, alignment, and susceptibility to adversarial manipulation. While prior benchmarks assess model refusal capabilities for harmful prompts, they often lack clinical specificity, graded harmfulness levels, and coverage of jailbreak-style attacks. We introduce CARES (Clinical Adversarial Robustness and Evaluation of Safety), a benchmark for evaluating LLM safety in healthcare. CARES includes over 18,000 prompts spanning eight medical safety principles, four harm levels, and four prompting styles: direct, indirect, obfuscated, and role-play, to simulate both malicious and benign use cases. We propose a three-way response evaluation protocol (Accept, Caution, Refuse) and a fine-grained Safety Score metric to assess model behavior. Our analysis reveals that many state-of-the-art LLMs remain vulnerable to jailbreaks that subtly rephrase harmful prompts, while also over-refusing safe but atypically phrased queries. Finally, we propose a mitigation strategy using a lightweight classifier to detect jailbreak attempts and steer models toward safer behavior via reminder-based conditioning. CARES provides a rigorous framework for testing and improving medical LLM safety under adversarial and ambiguous conditions.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. DataDignity: Training Data Attribution for Large Language Models

    cs.AI 2026-05 unverdicted novelty 7.0

    ScoringModel raises mean Recall@10 to 52.2 on the FakeWiki provenance benchmark from 35.0 for the best baseline, winning 41 of 45 model-by-condition comparisons and gaining 15.7 points on jailbreak-style queries.

  2. Asymmetric Collapse in Model Merging: When Refusal Over- writes Recognition

    cs.AI 2026-07 conditional novelty 6.0

    Across four standard merging methods, refusal behavior from a large task vector overwrites fine-grained harm classification, leaving at most 12.9% accuracy.

  3. Aligning Clinical Needs and AI Capabilities: A Survey on LLMs for Medical Reasoning

    cs.AI 2026-07 accept novelty 6.0

    A dual clinical-computational taxonomy for medical LLM reasoning plus a five-level 5k-sample benchmark showing specialists excel at diagnosis and general models at decision support/dialogue.

  4. LegalCiteBench: Evaluating Citation Reliability in Legal Language Models

    cs.CL 2026-05 unverdicted novelty 6.0

    LegalCiteBench reveals that current LLMs achieve under 7% accuracy on closed-book legal citation retrieval and completion tasks, with misleading answer rates above 94% for nearly all tested models.

  5. Chain of Risk: Safety Failures in Large Reasoning Models and Mitigation via Adaptive Multi-Principle Steering

    cs.AI 2026-05 unverdicted novelty 6.0

    Reasoning traces in large reasoning models expose safety failures missed by final-answer checks, and adaptive multi-principle steering reduces unsafe content in both traces and answers while preserving task performance.

  6. Understanding Annotator Safety Policy with Interpretability

    cs.AI 2026-05 unverdicted novelty 6.0

    Annotator Policy Models learn safety policies from labeling behavior alone, accurately predicting responses and revealing sources of disagreement like policy ambiguity and value pluralism.

  7. Dynamic Generation of Multi-LLM Agents Communication Topologies with Graph Diffusion Models

    cs.CL 2025-10 unverdicted novelty 6.0

    GTD generates task-adaptive, sparse communication topologies for multi-LLM agents via guided iterative graph diffusion steered by a proxy model predicting accuracy, utility, and cost.

  8. PolicyAlign: Direct Policy-Based Safety Alignment for Large Language Models

    cs.CL 2026-06 unverdicted novelty 5.0

    PolicyAlign aligns LLMs to natural-language safety policies by synthesizing violating instructions and performing on-policy self-distillation with policy-sensitive filtering, improving safety without high-quality supe...