aiqbench analysis
This research introduces CPB-Bench, a new benchmark designed to test large language models against realistic patient behaviors like contradictions and self-diagnosis rather than idealized queries. The study reveals that current medical LLMs consistently fail when encountering medically implausible or inconsistent input, highlighting critical gaps in safety protocols for clinical deployment. Furthermore, the evaluation of intervention strategies suggests that standard correction methods may introduce unnecessary noise without reliably resolving these complex behavioral challenges.
Beyond Idealized Patients: Evaluating LLMs under Challenging Patient Behaviors in Medical Consultations
arXiv:2603.29373v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used for medical consultation and health information support, where safety depends not only on medical knowledge but also on robust responses to unclear, inconsistent, or misleading patient input. However, most existing medical LLM evaluations assume idealized and well-posed patient questions, limiting their realism. We study challenging patient behaviors that commonly arise in real medical consultations and complicate safe clinical reasoning. We define four clinically grounded categories of such behaviors: information contradiction, factual inaccuracy, self-diagnosis, and care resistance. For each behavior, we specify concrete failure criteria that capture unsafe responses. Building on four existing medical dialogue datasets, we introduce CPB-Bench (Challenging Patient Behaviors Benchmark), a bilingual (English and Chinese) benchmark of multi-turn dialogues annotated for these behaviors. We find that although models perform well overall, they exhibit consistent behavior-specific failures, especially when handling contradictory or medically implausible patient information. We further evaluate four intervention strategies and find inconsistent improvements, with some interventions introducing unnecessary corrections.
Abstract:Large language models (LLMs) are increasingly used for medical consultation and health information support, where safety depends not only on medical knowledge but also on robust responses to unclear, inconsistent, or misleading patient input. However, most existing medical LLM evaluations assume idealized and well-posed patient questions, limiting their realism. We study challenging patient behaviors that commonly arise in real medical consultations and complicate safe clinical reasoning. We define four clinically grounded categories of such behaviors: information contradiction, factual inaccuracy, self-diagnosis, and care resistance. For each behavior, we specify concrete failure criteria that capture unsafe responses. Building on four existing medical dialogue datasets, we introduce CPB-Bench (Challenging Patient Behaviors Benchmark), a bilingual (English and Chinese) benchmark of multi-turn dialogues annotated for these behaviors. We find that although models perform well overall, they exhibit consistent behavior-specific failures, especially when handling contradictory or medically implausible patient information. We further evaluate four intervention strategies and find inconsistent improvements, with some interventions introducing unnecessary corrections.
Computation and Language (cs.CL)
arXiv:2603.29373 [cs.CL]
(or arXiv:2603.29373v2 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2603.29373 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 253 words
Related coverage
The Continual Constitutional Convention: How language models revise the documents that govern them
AI Alignment Forum · Oct 13, 2026
Amazon Research Awards recipients announced
Amazon Science · May 27, 2026
Concrete Evaluations to Investigate the OpenAI Model That Hacked Hugging Face
AI Alignment Forum · Aug 3, 2026
Deconfusing Corrigibility
AI Alignment Forum · Oct 17, 2026