Dynamic Triangulation-Based Graph Rewiring for Graph Neural Networks

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Attali, Hugo, Papastergiou, Thomas, Pernelle, Nathalie, Malliaros, Fragkiskos D.
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911126855352320
author Attali, Hugo
Papastergiou, Thomas
Pernelle, Nathalie
Malliaros, Fragkiskos D.
author_facet Attali, Hugo
Papastergiou, Thomas
Pernelle, Nathalie
Malliaros, Fragkiskos D.
contents Graph Neural Networks (GNNs) have emerged as the leading paradigm for learning over graph-structured data. However, their performance is limited by issues inherent to graph topology, most notably oversquashing and oversmoothing. Recent advances in graph rewiring aim to mitigate these limitations by modifying the graph topology to promote more effective information propagation. In this work, we introduce TRIGON, a novel framework that constructs enriched, non-planar triangulations by learning to select relevant triangles from multiple graph views. By jointly optimizing triangle selection and downstream classification performance, our method produces a rewired graph with markedly improved structural properties such as reduced diameter, increased spectral gap, and lower effective resistance compared to existing rewiring methods. Empirical results demonstrate that TRIGON outperforms state-of-the-art approaches on node classification tasks across a range of homophilic and heterophilic benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2508_19071
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dynamic Triangulation-Based Graph Rewiring for Graph Neural Networks
Attali, Hugo
Papastergiou, Thomas
Pernelle, Nathalie
Malliaros, Fragkiskos D.
Machine Learning
Artificial Intelligence
Graph Neural Networks (GNNs) have emerged as the leading paradigm for learning over graph-structured data. However, their performance is limited by issues inherent to graph topology, most notably oversquashing and oversmoothing. Recent advances in graph rewiring aim to mitigate these limitations by modifying the graph topology to promote more effective information propagation. In this work, we introduce TRIGON, a novel framework that constructs enriched, non-planar triangulations by learning to select relevant triangles from multiple graph views. By jointly optimizing triangle selection and downstream classification performance, our method produces a rewired graph with markedly improved structural properties such as reduced diameter, increased spectral gap, and lower effective resistance compared to existing rewiring methods. Empirical results demonstrate that TRIGON outperforms state-of-the-art approaches on node classification tasks across a range of homophilic and heterophilic benchmarks.
title Dynamic Triangulation-Based Graph Rewiring for Graph Neural Networks
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2508.19071