ExplainCPE
收藏资源简介:
ExplainCPE是一个针对中国药师考试的自由文本解释基准数据集,包含超过7000个问题,专门设计用于评估模型生成的解释。该数据集由哈尔滨工业大学(深圳)创建,旨在解决大型语言模型在医疗领域的解释性问题。数据集内容涵盖广泛的医疗知识,通过从官方考试解决方案中提取的解释来确保数据的准确性和完整性。ExplainCPE的应用领域主要集中在提高AI在医疗诊断中的解释能力,减少误诊,并推动智能医疗的发展。
ExplainCPE is a free-text explanatory benchmark dataset tailored for the Chinese Pharmacist Licensure Examination. Comprising over 7,000 questions, it is specifically designed to evaluate model-generated explanations. Developed by Harbin Institute of Technology (Shenzhen), this dataset aims to address the explanatory challenges faced by large language models in the medical domain. Covering a broad spectrum of medical knowledge, it ensures data accuracy and integrity through explanations extracted from official exam solutions. The primary application scenarios of ExplainCPE focus on enhancing the explanatory capability of AI in medical diagnosis, reducing misdiagnosis, and advancing the development of intelligent healthcare.




