Benchmark RadarAI BENCHMARK PROFILE
Endo-C6
Evaluates temporal vision-language models on surgical endoscopy video understanding under six realistic corruptions, using public videos and standardized prompts.
- Released
- 2026-08-14
- Readiness
- Paper only
- Primary field
- Health & Life Sciences
Why it matters
Provides a standardized robustness benchmark for clinical vision-language systems, exposing worst-case performance degradation under clinically relevant artifacts.
Motivation
Temporal vision-language models (TVLMs) offer a reusable, prompt-based interface for surgical video understanding, yet, their robustness under clinically realistic acquisition artifacts in endoscopy remains insufficiently characterized.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.