arXiv cs.AI· Yulin Fu (Beijing University of Posts and Telecommunications), Junren Wang (West China Hospital, Sichuan Provincial Engineering Research Center of Intelligent Diagnosis and Treatment of Breast Diseases), Guangjing Yang (Beijing University of Posts and Telecommunications), Zhangyuan Yu (Beijing University of Posts and Telecommunications), Wanran Sun (Beijing University of Posts and Telecommunications), Jiabao Zhou (Beijing University of Posts and Telecommunications), Jin Yin (West China Hospital, Sichuan Provincial Engineering Research Center of Intelligent Diagnosis and Treatment of Breast Diseases), Qicheng Lao (Beijing University of Posts and Telecommunications)·· 25 天前AI 评分35
MedBenchAgent：医疗 VLM 基准构建的系统化自动化框架
MedBenchAgent: Towards Systematic Automation of Medical VLM Benchmark Construction
AI 导读
研究人员提出多智能体框架 MedBenchAgent，将医疗 VLM 基准构建表述为受限编译过程，以系统化推导基准规范并生成评测项目。该框架采用基准中间表示（BIR）解耦规划与实例化阶段，取得 90.9% 的 Task-Space F1，抽样的 1,000 个项目中 994 个通过人工审核。该框架已成功迁移至专门医疗领域并评测了 12 个 VLM。
来源：arXiv cs.AI · arxiv.org