IatroBench: A Pre-Registered Benchmark of Clinical Omission in Language Models
AI 导读
临床遗漏基准测试框架依赖隐瞒模型表现差异数据缺失问题
IatroBench基准测试揭示语言模型在医疗场景中存在框架依赖的信息隐瞒现象，不同提问方式导致模型分享信息量差异显著。
IatroBench evaluates language models' clinical omission risks, revealing framing-dependent information withholding across medical scenarios.
重点速览
- IatroBench评估模型在60个预注册临床场景中的信息遗漏风险 IatroBench evaluates clinical omission risks across 60 pre-registered scenarios
- 不同框架（患者查询/医生咨询）导致模型分享信息量差异显著 Information sharing varies significantly between patient query and doctor consultation frameworks
- Claude Opus 4.6的隐瞒评分与医生评分高度一致 Claude Opus 4.6's omission scores align closely with physician assessments
- Llama 4在两种框架下均表现不佳，GPT-5.2数据缺失率达33.2% Llama 4 performs poorly in both framing contexts
- 标准LLM法官对遗漏危害的评分与结构化评估存在分歧 Standard LLM judges differ from structured evaluations on omission harm
一句话：模型在医疗场景中存在框架依赖的信息隐瞒问题 / Language models exhibit framing-dependent information withholding in clinical contexts
推荐理由 揭示模型在医疗场景中的信息控制机制，对安全应用有指导意义
二次创作声明：本页为 AI 热榜聚合导读，内容与热度数据来自公开来源 （arxiv），版权归原始作者所有；本站仅做转载指引与摘要评述，不复制原文。