MedClawBench

What is MedClawBench?

MedClawBench evaluates long-horizon temporal reasoning in surgical videos through 1,123 doctor-grounded questions over self-built long neurosurgery recordings…

Released

2026-08-14

Evaluates

Agents, General AI, Reasoning, Vision & 3D

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2608.14015

MedClawBench paper

• Paper https://arxiv.org/abs/2608.14015
• Website https://fyycs.github.io/medclaw/

Which sources cite MedClawBench?

• MedClawBench

Related Agents, General AI, Reasoning, Vision & 3D benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• 5-Dialects-BN
• A-OKVQA
• A3S-Bench
• AARRI-Bench
• ActBench
• ActiveFly-Bench