Research map / Video / Video-language understanding
SVAgent: Storyline-guided Long Video Understanding via Cross-Modal Multi-Agent Collaboration
Zhongyu Yang, Zuhao Yang, Shuo Zhan, Tan Yue, Wei Pang, Yingfang Yuan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add SVAgent multi-agent framework | Qwen2.5-VL-3B · LongVideoBench | accuracy: 53.0 -> 59.7 | Improved |
| add SVAgent | Qwen3-VL-4B · MLVU | accuracy: 53.4 -> 64.9 | Improved |
| add SVAgent | Qwen2.5-VL-7B · VideoMME medium | accuracy: 69.0 -> 75.8 | Improved |
| add SVAgent | Qwen2.5-VL-7B · VideoMME long | accuracy: 52.3 -> 57.3 | Improved |
| add storyline agent alone (ablation) | Qwen2.5-VL-7B · LongVideoBench and LVBench | accuracy: +5.6 LongVideoBench, +4.9 LVBench | Improved |
| add textual verification (ablation) | Qwen2.5-VL-7B · MLVU | accuracy: +3.8 | Improved |
| add visual verification (ablation) | Qwen2.5-VL-7B · VideoMME long | accuracy: +3.5 | Improved |
| add meta-decision agent (ablation) | Qwen2.5-VL-7B · LongVideoBench | accuracy: +3.0 | Improved |
| swap SigLIP for CLIP or LongCLIP as DPP retrieval encoder (ablation) | SVAgent · LongVideoBench | accuracy: 60.7 -> 60.2 (CLIP), 59.9 (LongCLIP) | No clear effect |
| raise intersection ratio tau (ablation) | SVAgent · Video-MME | accuracy: 64.1 -> 67.2; latency 11.8s -> 38.5s | Mixed |
| increase refinement iterations T up to 10 (ablation) | SVAgent · Video-MME | accuracy: accuracy up to 68.2; latency 24.9s -> 57.3s | Mixed |
| use SVAgent frame selection instead of uniform sampling at 8 frames | Qwen2.5-VL-7B · Video-MME | accuracy: uniform 59.8% at 32 frames vs SVAgent 60.7% at 8 frames | Improved |
Closest papers by meaning
- Think, Then Verify: A Hypothesis-Verification Multi-Agent Framework for Long Video Understanding (CVPR 2026 · similarity 0.90)
- VideoMind: A Chain-of-LoRA Agent for Temporal-Grounded Video Reasoning (ICLR 2026 · similarity 0.89)
- VideoAgent: Long-form Video Understanding with Large Language Model as Agent (ECCV 2024 · similarity 0.89)
- Searching Videos as Trees: Self-Correcting Agents for Grounded Long Video QA (NeurIPS 2026 · similarity 0.88)
- MoReVQA: Exploring Modular Reasoning Models for Video Question Answering (CVPR 2024 · similarity 0.88)
- Enhancing Video-LLM Reasoning via Agent-of-Thoughts Distillation (CVPR 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
