AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Are Common Substructures Transferable? Riemannian Graph Foundation Model with Neural Vector Bundles

Li Sun, Zhenhao Huang, Yiding Wang, Qin Chen, Pietro Lio, Philip S. Yu

ICML 2026 · Graph learning › Graph self-supervised learning · approach: Heterophily and imbalance on graphs · 0 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
use GAUGE (neural vector bundle + Dirichlet loss) instead of best prior GFMgraph foundation models · PubMed (1-shot)accuracy (%): 56.82 -> 61.26 (vs RiemannGFM)Improved
use GAUGEGraphAny · Photo (5-shot)accuracy (%): 78.45 -> 81.33Improved
use GAUGEGFT · Roman-empire (5-shot)accuracy (%): 26.60 -> 26.43No clear effect
use GAUGE for zero-shot link predictionbest baseline · Facebook zero-shot link predictionAUC (%): 87.41 -> 93.88 (vs RAGraph)Improved
use GAUGE for zero-shot link predictionGCOPE · PubMed zero-shot link predictionAP (%): 61.91 -> 61.40Hurt
use GAUGE for zero-shot link predictionSAMGPT · Photo zero-shot link predictionAUC (%): 92.26 -> 93.04Improved
use GAUGE for graph isomorphismSAMGPT · CSLaccuracy (%): 64.17 -> 92.56Improved
use GAUGE for graph isomorphismGIN · ZINC12KMAE: 0.163 -> 0.157Improved

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