MSJoE: Jointly Evolving MLLM and Sampler for Efficient Long-Form Video Understanding
This work addresses the challenge of efficient long-form video understanding for MLLMs, providing improved accuracy for researchers and practitioners working with video analysis.
This paper introduces MSJoE, a framework that jointly evolves a Multimodal Large Language Model (MLLM) and a lightweight key-frame sampler to efficiently understand long-form videos. MSJoE achieves an 8.0% accuracy gain over the base MLLM and 1.1% higher accuracy than the strongest baseline on various long-video QA benchmarks.
Efficiently understanding long-form videos remains a fundamental challenge for multimodal large language models (MLLMs). In this paper, we present MLLM-Sampler Joint Evolution (MSJoE), a novel framework that jointly evolves the MLLM and a lightweight key-frame sampler for efficient long-form video understanding. MSJoE builds upon a key assumption that only a small subset of key-frames is truly informative for answering each question to a video. Specifically, MSJoE first reasons out several queries, which describe diverse visual perspectives relevant to the question. Then, these queries interact with a frozen CLIP model to produce a query-frame similarity matrix. Finally, a lightweight sampler predicts key-frame sampling weights from this matrix, selecting a compact set of informative frames, which are then fed into the MLLM for answer generation. Both the MLLM and sampler are jointly optimized through reinforcement learning, enabling co-adaptation of query-reasoning, frame-sampling, and key-frame understanding. A new long-video QA dataset containing 2.8K videos with 7K question-answer pairs is collected to support the training process. Extensive experiments on VideoMME, LongVideoBench, LVBench, and MLVU show that MSJoE achieves 8.0\% accuracy gain upon the base MLLM, and 1.1\% higher accuracy than strongest baseline method.