Video-MMMU: Evaluating Knowledge Acquisition from Multi-Discipline Professional Videos
This addresses the problem of systematically assessing video-based knowledge acquisition in LMMs for AI research, though it is incremental as it builds on existing benchmarking efforts.
The authors tackled the lack of benchmarks for evaluating knowledge acquisition in Large Multimodal Models (LMMs) from videos, introducing Video-MMMU with 300 videos and 900 questions across six disciplines, and found a steep performance decline in LMMs as cognitive demands increased, with a significant gap compared to humans.
Humans acquire knowledge through three cognitive stages: perceiving information, comprehending knowledge, and adapting knowledge to solve novel problems. Videos serve as an effective medium for this learning process, facilitating a progression through these cognitive stages. However, existing video benchmarks fail to systematically evaluate the knowledge acquisition capabilities in Large Multimodal Models (LMMs). To address this gap, we introduce Video-MMMU, a multi-modal, multi-disciplinary benchmark designed to assess LMMs' ability to acquire and utilize knowledge from videos. Video-MMMU features a curated collection of 300 expert-level videos and 900 human-annotated questions across six disciplines, evaluating knowledge acquisition through stage-aligned question-answer pairs: Perception, Comprehension, and Adaptation. A proposed knowledge gain metric, Δknowledge, quantifies improvement in performance after video viewing. Evaluation of LMMs reveals a steep decline in performance as cognitive demands increase and highlights a significant gap between human and model knowledge acquisition, underscoring the need for methods to enhance LMMs' capability to learn and adapt from videos.