MedStreamBench
Explore score history ↗MedStreamBench is a time-aware benchmark for medical video understanding, integrating 22 medical datasets and 5,419 QA instances across four temporal settings: retrospective, present, future, and proactive. Models are restricted to temporally bounded evidence windows and evaluated on answer correctness, responsiveness, and post-evidence stability.
- Released
- 2026-07-02
- Institutions
- Zhejiang University · Shanghai Key Laboratory of Trustworthy Computing · Carnegie Mellon University
- Evaluates
- Biology & Drug Discovery, Vision & 3D, cs.CV
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- unreviewed
MedStreamBench paper and dataset
No reported scores are on record for this benchmark yet.