MedQA-CS
收藏资源简介:
MedQA-CS数据集是由马萨诸塞大学阿默斯特分校等机构创建的,旨在评估大型语言模型在临床技能方面的表现。该数据集包含1667个数据点,涵盖信息收集、体格检查、诊断和治疗等多个临床场景。数据集的创建过程严格遵循USMLE Step 2 CS指南,并通过专家注释确保数据质量。MedQA-CS主要用于评估和提升AI在医疗领域的临床能力,特别是在复杂临床情境下的表现。
MedQA-CS dataset was developed by the University of Massachusetts Amherst and other institutions, aiming to evaluate the performance of large language models (LLMs) in clinical skills. This dataset comprises 1,667 data points, covering multiple clinical scenarios including information gathering, physical examination, diagnosis, and treatment. Its development strictly adheres to the USMLE Step 2 CS guidelines, and data quality is ensured through expert annotations. MedQA-CS is primarily used to assess and enhance the clinical capabilities of AI in the healthcare field, especially its performance in complex clinical contexts.




