ASEHybrid: When Geometry Matters Beyond Homophily in Graph Neural Networks

TL;DR

This paper introduces ASEHybrid, a geometry-aware GNN architecture that leverages label informativeness and graph curvature to improve performance on heterophilous graphs where structure provides label-relevant information.

Contribution

It develops a theoretical framework connecting curvature-guided rewiring and positional geometry through label informativeness, and proposes a practical architecture, ASEHybrid, with proven theoretical guarantees.

Findings

ASEHybrid improves performance on label-informative heterophilous benchmarks.

Degree-based Forman curvature reshapes information flow without increasing expressivity.

Theoretical analysis relates curvature and label informativeness to spectral behavior of label signals.

Abstract

Standard message-passing graph neural networks (GNNs) often struggle on graphs with low homophily, yet homophily alone does not explain this behavior, as graphs with similar homophily levels can exhibit markedly different performance and some heterophilous graphs remain easy for vanilla GCNs. Recent work suggests that label informativeness (LI), the mutual information between labels of adjacent nodes, provides a more faithful characterization of when graph structure is useful. In this work, we develop a unified theoretical framework that connects curvature-guided rewiring and positional geometry through the lens of label informativeness, and instantiate it in a practical geometry-aware architecture, ASEHybrid. Our analysis provides a necessary-and-sufficient characterization of when geometry-aware GNNs can improve over feature-only baselines: such gains are possible if and only if graph structure carries label-relevant information beyond node features. Theoretically, we relate adjusted homophily and label informativeness to the spectral behavior of label signals under Laplacian smoothing, show that degree-based Forman curvature does not increase expressivity beyond the one-dimensional Weisfeiler--Lehman test but instead reshapes information flow, and establish convergence and Lipschitz stability guarantees for a curvature-guided rewiring process. Empirically, we instantiate ASEHybrid using Forman curvature and Laplacian positional encodings and conduct controlled ablations on Chameleon, Squirrel, Texas, Tolokers, and Minesweeper, observing gains precisely on label-informative heterophilous benchmarks where graph structure provides label-relevant information beyond node features, and no meaningful improvement in high-baseline regimes.