LongVideoAgent: Multi-Agent Reasoning with Long Videos

TL;DR

This paper introduces LongVideoAgent, a multi-agent system that enhances reasoning over hour-long videos by localizing relevant segments and extracting visual details, significantly improving performance on new episode-level datasets.

Contribution

The paper presents a novel multi-agent framework with reinforcement learning for long-video question answering, addressing limitations of prior methods that rely on summaries or limited tools.

Findings

Outperforms non-agent baselines on LongTVQA and LongTVQA+ datasets.

Reinforcement learning improves reasoning and planning capabilities.

Provides interpretable reasoning trajectories.

Abstract

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/.