MLVU-M

MLVU-M is a variant of the Multi-task Long Video Understanding Benchmark (MLVU) specifically designed for multimodal evaluation. It extends the core MLVU framework to assess models' ability to understand and reason about long video content through multimodal inputs, testing advanced video comprehension capabilities across different video genres and temporal contexts while emphasizing multimodal integration and reasoning.