AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

VKG-QA: Visual Knowledge Graph-based Question Answer for Large Multimodal Models

Yuntao Du, Yiming Wang, Renshuo Yuan, Jincheng Yue, Yijing Chen, Yue Fan, Bo Zhang, Qian Li, Lizhen Cui

CVPR 2026 · Large language models › Retrieval and knowledge · approach: Knowledge-augmented 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
present the knowledge graph as an image to an LMM instead of text triples to an LLMGPT-5 vs GPT-5-chat · VKG-QA text-vs-image subsetaverage accuracy: 77.1 -> 88.8Improved
visual input instead of text for connectivity assessmentGPT-5 · VKG-QA connectivityaccuracy: 56.6 -> 94.6Improved
visual input instead of text for multi-hop and superlative reasoningLMMs vs LLMs · VKG-QA multi-hop / superlativeaccuracyHurt
increase reasoning depth from 1-hop to 3-hop (ablation)GPT-5 · VKG-QA hopsaccuracy: 91.7 -> 80.2Hurt
scale Qwen2.5-VL from 7B to 72B (ablation)Qwen2.5-VL · VKG-QA general image comprehensionaccuracy: 62.8 -> 65.3Improved
scale Qwen2.5-VL from 7B to 32B and 72B (ablation)Qwen2.5-VL · VKG-QA graph-specific comprehension and reasoningaccuracyImproved
use closed-source rather than open-source LMMsLMMs · VKG-QAaverage accuracy: 85.6 -> 63.1 (best closed vs best open)Improved

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