arXiv · AI· Rui Hua, Yu Wei, Zixin Shu, Kai Chang, Dengying Yan, Jianan Xia, Hui Zhu, Xiaodong Li, Dongmei Jia, Zhuye Gao, Yanyan Meng, Haibin Yu, Fei Dong, Kan Ding, Runshun Zhang, Jian Yu, Jiansheng Li, Baoyan Liu, Wenjia Wang, Xuezhong Zhou·· 2026-10-15AI 评分29
LingLanMedBench:中医大语言模型统一评测基准
LingLanMedBench: A Unified Benchmark for Evaluating Large Language Models in Traditional Chinese Medicine
AI 导读
LingLanMedBench 涵盖 5 个领域、13 个子任务共 25,620 条数据,对 14 个大语言模型进行零样本评测。 结果显示全量基准最高分降至硬性子集的 31.9 分,暴露出临床推理缺陷。该基准为中医 LLM 研发提供标准化评估框架。
来源:arXiv · AI · arxiv.org