AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Aligning Vision to Language: Annotation-Free Multimodal Knowledge Graph Construction for Enhanced LLMs Reasoning

Junming Liu, Siyuan Meng, Yanting Gao, Song Mao, Pinlong Cai, Guohang Yan, Yirong Chen, Zilin Bian, Ding Wang, Botian Shi

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Instruction-tuned multimodal assistants · 31 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 VaLiK image-only knowledge graph to Qwen2.5-7BQwen2.5-7B · CrisisMMDaccuracy (avg across tasks): +4.41%Improved
add VaLiK text-image knowledge graph to Qwen2.5-7BQwen2.5-7B · CrisisMMDaccuracy (avg across tasks): +4.90%Improved
add VaLiK graph to larger Qwen2.5-32B/72BQwen2.5-32B/72B · CrisisMMDaccuracy: +2.0% to 2.5%Improved
add VaLiK KG to VLMs at inferenceLLaVA-34B · CrisisMMDaccuracy: +2.41% image-only, +3.59% text-imageImproved
add CoE-based VLM description generation (ablation)VaLiK · ScienceQA average (Text-Image KG)accuracy (%): 74.72 -> 80.57Improved
add cross-modal similarity verification pruning (ablation)VaLiK · CrisisMMD Task 2 (Image-only KG)accuracy (%): 47.00 -> 49.54Improved
add similarity verification on ScienceQA image-only natural science (ablation)VaLiK · ScienceQA NAT (Image-only KG)accuracy (%): 80.06 -> 79.14Hurt
add similarity verification on ScienceQA overall average (Text-Image KG) (ablation)VaLiK · ScienceQA average (Text-Image KG)accuracy (%): 80.57 -> 83.16Improved
use VaLiK MMKG instead of pre-built Visual Genome or MmkgQwen2.5-7B · ScienceQA averageaccuracy (%): 75.08 (VG) -> 83.16Improved
use Mmkg as external knowledge for Qwen2.5-7BQwen2.5-7B · ScienceQA averageaccuracy (%): 74.72 -> 73.47Hurt
add more VLMs to the CoE cascade (ablation)VaLiK · CrisisMMDaccuracyNo clear effect
use larger retrieval model (32B/72B) for LightRAG retrieval (ablation)VaLiK · retrievalperformance: 1%-5% improvementImproved

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