Blinded physician evaluation has been considered by many to be the gold standard for assessing clinical reasoning in large language models (LLMs). This is di…
机构:哈佛医学院
来源:arXiv 2609.12822 | AI4Papers 论文推荐平台