AI Paper Atlas

Research map / Video / Video-language understanding

Symphony: A Cognitively-Inspired Multi-Agent System for Long-Video Understanding

Haiyang Yan, Hongyun Zhou, Peng Xu, Xiaoxue Feng, Mengyi Liu

CVPR 2026 · Video › Video-language understanding · approach: Long video understanding · 4 citations (OpenAlex)

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 changedWhereMeasuredEffect
add a separate reflection agent instead of planner self-reflection (ablation)Symphony · LVBenchaccuracy (%): -2.5% when removedImproved
add a subtitle agent instead of feeding full subtitles to the planner (ablation)Symphony · LVBenchaccuracy (%): -1.4% when removedImproved
add a specialized visual perception agent instead of merging its tools into the planner (ablation)Symphony · LVBenchaccuracy (%): +2.2%Improved
use the full multi-agent system instead of the planning agent alone (ablation)Symphony · LVBenchaccuracy (%): 65.7 -> 71.8Improved
use VLM-based scoring grounding with CLIP fallback instead of CLIP-only retrieval (ablation)Grounding agent · LVBenchaccuracy (%): 52.2 -> 71.8Improved
always use VLM scoring (no CLIP) instead of adaptive selection (ablation)Grounding agent · LVBenchaccuracy (%): 71.8 -> 72.1, time 54.8 -> 68.6 sMixed
use Seed 1.6VL instead of Qwen2.5VL-7B as scoring VLM (ablation)Grounding agent · LVBenchaccuracy (%): 68.6 -> 71.8Improved
lower sampling FPS from 0.5 to 0.25 (ablation)Grounding agent · LVBench EU subsetaccuracy (%): 68.6 -> 62.1Hurt
extend clip interval from 60 s to 120 s (ablation)Grounding agent · LVBench EU subsetaccuracy (%): 68.6 -> 64.7Hurt
increase FPS to 1 or shorten clip to 30 s (ablation)Grounding agent · LVBench EU subsetaccuracy (%): 68.6 -> 70.9 (FPS 1), 70.3 (30 s clip)No clear effect
add 3-instance majority voting (ablation)Symphony · LVBenchaccuracy (%): 71.8 -> 73.7Improved
add 3-instance majority voting (ablation)Symphony · MLVUaccuracy (%): 81.0 -> 83.6Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy