AI Paper Atlas

Research map / Vision-language and multimodal / Visual question answering

M^3KG-RAG: Multi-hop Multimodal Knowledge Graph-enhanced Retrieval-Augmented Generation

Hyeongcheol Park, Jiyoung Seo, Jaewon Mun, Hogun Park, Wonmin Byeon, Sung June Kim, Hyeonsoo Im, JeungSub Lee, Sangpil Kim

CVPR 2026 · Vision-language and multimodal › Visual question answering · approach: Multimodal LLM question answering

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 modality-wise retrieval over the multi-hop MMKG (ablation)M3KG-RAG with Qwen2.5-Omni · VALORModel-as-Judge score: 36.62 -> 40.91Improved
add GRASP on a text-only KG without modality-wise retrieval (ablation)M3KG-RAG with Qwen2.5-Omni · VALORModel-as-Judge score: 36.62 -> 36.96No clear effect
combine modality-wise retrieval with GRASP (ablation)M3KG-RAG with Qwen2.5-Omni · VALORModel-as-Judge score: 40.91 -> 44.67Improved
add GroundingDINO visual grounding to GRASP (ablation)M3KG-RAG with Qwen2.5-Omni · VALORModel-as-Judge score: 40.91 -> 41.35Improved
add TAG audio grounding to GRASP (ablation)M3KG-RAG with Qwen2.5-Omni · VALORModel-as-Judge score: 40.91 -> 41.70Improved
use both visual and audio grounding in GRASP (ablation)M3KG-RAG with Qwen2.5-Omni · VALORModel-as-Judge score: 40.91 -> 42.96Improved
add LLM-based filtering stage to GRASP (ablation)M3KG-RAG with Qwen2.5-Omni · VALORModel-as-Judge score: 42.96 -> 44.87Improved
LLM filter increases VRAM and latency (ablation)GRASP · VALORGPU VRAM (GB): 24.2 -> 39.8Hurt
use Wikidata text KG with naive RAG instead of no retrievalQwen2.5-Omni · VALORModel-as-Judge score: 32.42 -> 30.28Hurt
use VAT-KG single-hop MMKG instead of no retrievalQwen2.5-Omni · VALORModel-as-Judge score: 32.42 -> 35.44Improved
set distance threshold tau too small or too large (ablation)M3KG-RAG · VALORModel-as-Judge scoreHurt
raise GRASP presence threshold eta_av from 1.2 to 1.5 and 1.8 (ablation)M3KG-RAG · VALORModel-as-Judge scoreHurt

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