Benchmark RadarAI BENCHMARK PROFILE
Ego-MC-Bench
Ego-MC-Bench is a benchmark for evaluating reactive step-by-step task guidance in cooking scenarios, focusing on timely interventions when mistakes occur.
- Released
- 2026-06-08
- Readiness
- Inspectable
- Primary field
- General AI
Why it matters
The benchmark addresses the capability of video LLMs to intervene proactively during task execution, which is crucial for practical guidance assistants.
Motivation
Learning everyday skills, like cooking a dish, relies increasingly on instructional media such as online videos.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.