AI Paper Atlas

Research map / Video / Video-language understanding

ViG-RAG: Video-aware Graph Retrieval-Augmented Generation via Temporal and Semantic Hybrid Reasoning

Zongsheng Cao, Anran Liu, Yangfan He, Jing Li, Bo Zhang, Zigan Wang

AAAI 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained 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
add ViG-RAG retrieval to Video-LLaVAVideo-LLaVA 7B · Video-MMEoverall accuracy: 39.6 -> 43.5Improved
add ViG-RAG retrieval to LLaVA-NeXT-VideoLLaVA-NeXT-Video 7B · Video-MMEoverall accuracy: 43.0 -> 54.2Improved
add ViG-RAG retrieval to LongVALongVA 7B · Video-MMEoverall accuracy: 51.4 -> 58.8Improved
add ViG-RAG retrieval to Long-LLaVALong-LLaVA 7B · Video-MMEoverall accuracy: 52.0 -> 57.7Improved
add ViG-RAG retrieval to Qwen2-VLQwen2-VL 72B · Video-MMEoverall accuracy: 64.9 -> 72.4Improved
add ViG-RAG retrieval to LLaVA-VideoLLaVA-Video 72B · Video-MMEoverall accuracy: 67.1 -> 74.4Improved
add ViG-RAG retrieval to Qwen2-VL, long videosQwen2-VL 72B · Video-MME (long)accuracy: 56.3 -> 72.2Improved
ViG-RAG instead of VideoRAG in win-rate comparison vs GraphRAG-lViG-RAG · LongerVideos (overall)Overall win rate: 35.81% -> 41.77%Improved
add probabilistic temporal knowledge graph (PTKG) (ablation)ViG-RAG · benchmarks (Figure 3)retrieval accuracy / overall performanceImproved
add query enhancement (ablation)ViG-RAG · benchmarks (Figure 3)retrieval accuracy / overall performanceImproved
add GMM-based adaptive filtering (ablation)ViG-RAG · benchmarks (Figure 3)retrieval accuracy / overall performanceImproved

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