# SVAgent: Storyline-Guided Long Video Understanding via Cross-Modal Multi-Agent Collaboration Source: [SVAgent: Storyline-Guided Long Video Understanding via Cross-Modal Multi-Agent Collaboration]() Collection index: [llms.txt]() Data updated: 2026-04-09T05:00:04.534Z Authors: Zhongyu Yang, Zuhao Yang, Shuo Zhan, Tan Yue, Wei Pang, Yingfang Yuan Published: 2026-04-06T18:30:50.000Z ## Abstract Video question answering (VideoQA) is a challenging task that requires integrating spatial, temporal, and semantic information to capture the complex dynamics of video sequences. Although recent advances have introduced various approaches for video understanding, most existing methods still rely on locating relevant frames to answer questions rather than reasoning through the evolving storyline as humans do. Humans naturally interpret videos through coherent storylines, an ability that is crucial for making robust and contextually grounded predictions. To address this gap, we propose SVAgent, a storyline-guided cross-modal multi-agent framework for VideoQA. The storyline agent progressively constructs a narrative representation based on frames suggested by a refinement suggestion agent that analyzes historical failures. In addition, cross-modal decision agents independently predict answers from visual and textual modalities under the guidance of the evolving storyline. Their outputs are then evaluated by a meta-agent to align cross-modal predictions and enhance reasoning robustness and answer consistency. Experimental results demonstrate that SVAgent achieves superior performance and interpretability by emulating human-like storyline reasoning in video understanding. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Video Question Answering on LVBench (test)](): Video Question Answering; LVBench (test); Accuracy: 40.8. - [Video Question Answering on MLVU (test)](): Video Question Answering; MLVU (test); Accuracy: 65.6. - [Video Question Answering on LongVideoBench w/o sub. (val)](): Video Question Answering; LongVideoBench w/o sub. (val); Accuracy: 61. - [Video Question Answering on VideoMME (Length and Subtitle Breakdown)](): Video Question Answering; VideoMME; Accuracy (Overall, w/o Subtitles): 63.1. ## Original sources and implementation [Original paper]() [PDF]()