LongVideoAgent: Multi-Agent Reasoning with Long Videos

About this episode

🤗 Upvotes: 38 | cs.AI, cs.CV, cs.LG, cs.MA

        <strong>Authors:</strong><br />
        Runtao Liu, Ziyi Liu, Jiaqi Tang, Yue Ma, Renjie Pi, Jipeng Zhang, Qifeng Chen</p>

        <strong>Title:</strong><br />
        LongVideoAgent: Multi-Agent Reasoning with Long Videos</p>

        <strong>Arxiv:</strong><br />
        <a href="http://arxiv.org/abs/2512.20618v1">http://arxiv.org/abs/2512.20618v1</a></p>

        <strong>Abstract:</strong><br />
        Recent advances in multimodal LLMs and systems that use tools for long-video QA point to the promise of reasoning over hour-long episodes. However, many methods still compress content into lossy summaries or rely on limited toolsets, weakening temporal grounding and missing fine-grained cues. We propose a multi-agent framework in which a master LLM coordinates a grounding agent to localize question-relevant segments and a vision agent to extract targeted textual observations. The master agent plans with a step limit, and is trained with reinforcement learning to encourage concise, correct, and efficient multi-agent cooperation. This design helps the master agent focus on relevant clips via grounding, complements subtitles with visual detail, and yields interpretable trajectories. On our proposed LongTVQA and LongTVQA+ which are episode-level datasets aggregated from TVQA/TVQA+, our multi-agent system significantly outperforms strong non-agent baselines. Experiments also show reinforcement learning further strengthens reasoning and planning for the trained agent. Code and data will be shared at https://longvideoagent.github.io/