collaborators

1 paper

cs.CV2026

MedOpenClaw and MedFlowBench: Auditing Medical Agents in Full-Study Workflows

Weixiang Shen, Chengzhi Shen, Yanzhu Hu +12

Medical imaging benchmarks often evaluate VLMs on pre-selected 2D images, slices, crops, or patches, making evaluation closer to visual recognition. Real clinical workflows impose…