AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Learning Pseudo-Riemannian Manifolds for Heterophilic Graphs via Graph Signature

Yun Young Choi, Asung Kil, Sun Woo Park, Minho Lee, Seokhwan Kim

NeurIPS 2026 · Graph learning › Graph self-supervised learning · approach: Semi-supervised node representation learning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes learning pseudo-Riemannian manifolds (indefinite metric signatures) for graph neural networks to better model heterophilic graphs, where edges signify dissimilarity.

Problem

GNNs implicitly assume a Riemannian (positive-definite metric) data manifold, creating a geometric mismatch for heterophilic graphs where edges often encode dissimilarity or structural repulsion.

Approach

Shift from Riemannian to pseudo-Riemannian geometry, learning the metric signature of the manifold via a graph signature so that the geometry can represent repulsive (dissimilar) relations alongside attractive ones.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
allow indefinite metric (vs positive-definite Lambda) (ablation)PRAT · 8 standard benchmarks, averageaccuracy: 68.29 -> 71.96Improved
allow indefinite metric (vs positive-definite Lambda) (ablation)PRAT · Cornellaccuracy: 57.58 -> 71.89Improved
allow indefinite metric (vs positive-definite Lambda) (ablation)PRAT · Texasaccuracy: 76.78 -> 84.32Improved
allow indefinite metric (vs positive-definite Lambda) (ablation)PRAT · Chameleonaccuracy: 65.64 -> 68.73Improved
allow indefinite metric (vs positive-definite Lambda) (ablation)PRAT · Cora, PubMed, Citeseeraccuracy: Cora 88.42 vs 88.18; Citeseer 78.06 vs 78.09No clear effect
learn Lambda instead of freezing at signature initialization (ablation)PRAT · Cornellaccuracy: 55.15 -> 71.89Improved
learn Lambda instead of freezing at signature initialization (ablation)PRAT · Texasaccuracy: 74.89 -> 84.32Improved
learn Lambda instead of freezing it (ablation)PRAT · large Platonov graphsaccuracyNo clear effect
indefinite metric vs signed aggregation weights only (FAGCN) on synthetic graphsPRAT · synthetic graphs, h_E <= 0.4accuracy: gap averages 40.6 pointsImproved
PRAT vs GAT and GATv2 under varying graph signature at h_E=0.5PRAT · synthetic graphsaccuracy std across signatures: GAT 3.99, GATv2 3.42, PRAT 0.68Improved
delete negative directions of signature but keep magnitudes (ablation)PRAT · roman-empireaccuracy: -2.26 pointsHurt
replace signature with sign-only, sign-flipped or random draws (ablation)PRAT · roman-empireaccuracy: within 0.7No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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