MedExpQA

MedExpQA是由HiTZ中心和巴斯克大学创建的多语言医疗问答基准数据集，包含622条数据，涵盖英语、法语、意大利语和西班牙语。该数据集基于医疗考试，首次包含了由医生编写的参考黄金解释，用于评估大型语言模型在医疗问答任务中的表现。数据集旨在解决现有医疗知识评估基准的不足，特别是缺乏黄金解释和多语言支持的问题。MedExpQA的应用领域包括医疗决策支持和医疗教育，旨在通过提供高质量的医疗知识解释来提高AI在医疗领域的应用效果。

MedExpQA is a multilingual medical question answering benchmark dataset developed by HiTZ Center and the University of the Basque Country, consisting of 622 entries covering English, French, Italian and Spanish. Rooted in medical examinations, this dataset is the first to include gold standard reference explanations authored by physicians, and is designed to evaluate the performance of large language models (LLMs) on medical question answering tasks. It aims to address the shortcomings of existing medical knowledge evaluation benchmarks, particularly the lack of gold standard explanations and multilingual support. The application fields of MedExpQA include medical decision support and medical education, with the goal of improving the application effectiveness of AI in the healthcare domain by providing high-quality medical knowledge explanations.