MedGRPO Demo — Medical Video Understanding

This demo showcases MedGRPO fine-tuned on MedVidBench for medical video question answering across 8 tasks: temporal reasoning, spatial grounding, captioning, and clinical assessment.

📄 Paper   🌐 Project Page   💾 Dataset   🤖 Model   💻 GitHub   📊 Leaderboard

Browse pre-computed predictions from the test set (no GPU needed).

Select Task

Task

⏱️ Temporal Action Localization (TAL)

Identify when specific surgical actions occur in the video (start–end times).

Choose Example