AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

OpenMAG: A Comprehensive Benchmark for Multimodal-Attributed Graph

Chenxi Wan, Xunkai Li, Yilong Zuo, Haokun Deng, Sihan Li, Bowen Fan, Hongchao Qin, Rong-Hua Li, Guoren Wang

ICML 2026 · Graph learning › Graph self-supervised learning · approach: LLMs and graph foundation models · 0 citations (OpenAlex)

Read the paper

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
use multimodal features instead of unimodal text or visual features (ablation)MAG learning models · node classification and link prediction across datasetsaccuracy / MRRImproved
add graph neighbor aggregation (Graph-QFormer, 4 neighbors) to G2Image (ablation)G2Image on SemArt · SemArtDINO-Score: 50.86 -> 55.29Improved
increase aggregated neighbors from 4 to 8 (ablation)G2Image on SemArt · SemArtCLIP-Score: 70.21 -> 67.10Hurt
add graph neighbor aggregation (6 neighbors) to G2Image (ablation)G2Image on Toys · ToysDINO-Score: 28.13 -> 30.83Improved
fine-tune text and visual encoders end to end instead of freezing (ablation)MLP on node classification · Movies and Grocerytest accuracyImproved
use unified encoders (CLIP, Qwen2.5) instead of independent text plus image encoders (ablation)MLP on modality retrieval · Flickr30kretrieval Hits@K and MRRImproved
use graph-enhanced models (LGMRec, DGF) instead of MLP and traditional GNNsgraph-based tasks · Movies, DY link predictionaccuracy / MRR: 62.57 -> 77.28 MRR (MLP to DGF on DY)Improved
use MLLM-style generative masked-prediction model (UniGraph2) instead of discriminative modelsgraph-based tasks · Movies, Grocery, Toysaccuracy / NMI: 51.22 -> 46.78 acc (MLP to UniGraph2 on Movies)Hurt
add graph structure aggregation to MLP for modality retrievalMLP on modality retrieval · ToysT2I MRR: 99.46 -> 94.20 (DGF)Hurt
add LoRA fine-tuning to the LLM (ablation)OPT on G2Text · Flickr30kCIDEr: 41.51 -> 46.23Improved
add Adapter fine-tuning to the LLM (ablation)OPT on G2Text · Flickr30kCIDEr: 41.51 -> 34.95Hurt
add LoRA fine-tuning to the LLM (ablation)LLaMA on G2Text · Flickr30kCIDEr: 44.58 -> 47.76Improved

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