AI Paper Atlas

Research map / Video / Video-language understanding

Graph-to-Frame RAG: Visual-Space Knowledge Fusion for Training-Free and Auditable Video Reasoning

Songyuan Yang, Weijiang Yu, Ziyu Liu, Guijian Tang, Wenjing Yang, Huibin Tan, Nong Xiao

CVPR 2026 · Video › Video-language understanding · approach: Long video understanding · 3 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
deliver retrieved subgraph as a rendered reasoning frame instead of textual JSON (ablation)G2F-RAG on Qwen2.5-VL-7B · VideoMME (w/o sub)accuracy (%): 63.0 -> 70.6Improved
deliver retrieved subgraph as a rendered reasoning frame instead of textual JSON (ablation)G2F-RAG on Qwen2.5-VL-7B · WildVideoaccuracy (%): 48.8 -> 55.4Improved
append the reasoning frame at the end rather than mid-sequence (ablation)G2F-RAG on Qwen2.5-VL-7B · VideoMME (w/o sub)accuracy (%): 64.0 -> 70.6Improved
append the reasoning frame at the end rather than the start (ablation)G2F-RAG on Qwen2.5-VL-7B · VideoMME (w/o sub)accuracy (%): 69.4 -> 70.6Improved
use a single end frame instead of four end frames (ablation)G2F-RAG on Qwen2.5-VL-7B · MLVUaccuracy (%): 69.0 -> 73.4Improved
use a single end frame instead of two end frames (ablation)G2F-RAG on Qwen2.5-VL-7B · VideoMME (w/o sub)accuracy (%): 69.8 -> 70.6Improved
use minimal icon-and-short-label frame style instead of dense text (ablation)G2F-RAG on Qwen2.5-VL-7B · WildVideoaccuracy (%): 50.5 -> 55.4Improved
keep both spatial topology and causality instead of topology only (ablation)G2F-RAG on Qwen2.5-VL-7B · MLVUaccuracy (%): 71.0 -> 73.4Improved
add difficulty routing instead of applying the frame to all queries (ablation)G2F-RAG on Qwen2.5-VL-7B · VideoMME (w/o sub)accuracy (%): 66.8 -> 70.6Improved
add conservative fallback to routing (ablation)G2F-RAG on Qwen2.5-VL-7B · VideoMME (w/o sub)accuracy (%): 69.0 -> 70.6Improved
include intent and affordances in the offline graph (ablation)G2F-RAG on Qwen2.5-VL-7B · MLVUaccuracy (%): 70.2 -> 73.4Improved
enable external tools during graph construction (ablation)G2F-RAG on Qwen2.5-VL-7B · WildVideoaccuracy (%): 52.4 -> 55.4Improved

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