Enregistré dans:
Détails bibliographiques
Auteurs principaux: Huang, Sujia, Fu, Lele, Cui, Zhen, Zhang, Tong, Song, Na, Huang, Bo
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2507.21422
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918145364590592
author Huang, Sujia
Fu, Lele
Cui, Zhen
Zhang, Tong
Song, Na
Huang, Bo
author_facet Huang, Sujia
Fu, Lele
Cui, Zhen
Zhang, Tong
Song, Na
Huang, Bo
contents Graph Neural Networks (GNNs) have emerged as powerful tools for learning from graph-structured data, leveraging message passing to diffuse information and update node representations. However, most efforts have suggested that native interactions encoded in the graph may not be friendly for this process, motivating the development of graph rewiring methods. In this work, we propose a torque-driven hierarchical rewiring strategy, inspired by the notion of torque in classical mechanics, dynamically modulating message passing to improve representation learning in heterophilous and homophilous graphs. Specifically, we define the torque by treating the feature distance as a lever arm vector and the neighbor feature as a force vector weighted by the homophily disparity between nodes. We use the metric to hierarchically reconfigure receptive field of each layer by judiciously pruning high-torque edges and adding low-torque links, suppressing the impact of irrelevant information and boosting pertinent signals during message passing. Extensive evaluations on benchmark datasets show that the proposed approach surpasses state-of-the-art rewiring methods on both heterophilous and homophilous graphs.
format Preprint
id arxiv_https___arxiv_org_abs_2507_21422
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle GraphTorque: Torque-Driven Rewiring Graph Neural Network
Huang, Sujia
Fu, Lele
Cui, Zhen
Zhang, Tong
Song, Na
Huang, Bo
Machine Learning
Graph Neural Networks (GNNs) have emerged as powerful tools for learning from graph-structured data, leveraging message passing to diffuse information and update node representations. However, most efforts have suggested that native interactions encoded in the graph may not be friendly for this process, motivating the development of graph rewiring methods. In this work, we propose a torque-driven hierarchical rewiring strategy, inspired by the notion of torque in classical mechanics, dynamically modulating message passing to improve representation learning in heterophilous and homophilous graphs. Specifically, we define the torque by treating the feature distance as a lever arm vector and the neighbor feature as a force vector weighted by the homophily disparity between nodes. We use the metric to hierarchically reconfigure receptive field of each layer by judiciously pruning high-torque edges and adding low-torque links, suppressing the impact of irrelevant information and boosting pertinent signals during message passing. Extensive evaluations on benchmark datasets show that the proposed approach surpasses state-of-the-art rewiring methods on both heterophilous and homophilous graphs.
title GraphTorque: Torque-Driven Rewiring Graph Neural Network
topic Machine Learning
url https://arxiv.org/abs/2507.21422