ModelRefs / MLVU Methodology and Evidence
MLVU Methodology and Evidence
Multi-task long video understanding (3 min – 2 hr clips). Methodology, dataset provenance, and citation sources for MLVU.
Overview
Multi-task long video understanding (3 min – 2 hr clips).
How it is measured: M-Avg score across 9 tasks.
How this benchmark is scored
| Category | multimodal |
|---|---|
| Maximum score | 100 M-Avg |
| Direction | Higher is better |
Primary source: https://github.com/JUNJIE99/MLVU
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to MLVU Methodology and Evidence.
Frequently asked questions
What does the MLVU benchmark measure?
Multi-task long video understanding (3 min – 2 hr clips).
What is the methodology?
M-Avg score across 9 tasks.