Decoupling and Damping: Structurally-Regularized Gradient Matching for Multimodal Graph Condensation

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Shen, Lian, Chen, Zhendan, Song, Meijia, jiang, Yinhui, Su, Ziming, Liu, Juan, Liu, Xiangrong
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915785571565568
author Shen, Lian
Chen, Zhendan
Song, Meijia
jiang, Yinhui
Su, Ziming
Liu, Juan
Liu, Xiangrong
author_facet Shen, Lian
Chen, Zhendan
Song, Meijia
jiang, Yinhui
Su, Ziming
Liu, Juan
Liu, Xiangrong
contents In multimodal graph learning, graph structures that integrate information from multiple sources, such as vision and text, can more comprehensively model complex entity relationships. However, the continuous growth of their data scale poses a significant computational bottleneck for training. Graph condensation methods provide a feasible path forward by synthesizing compact and representative datasets. Nevertheless, existing condensation approaches generally suffer from performance limitations in multimodal scenarios, mainly due to two reasons: (1) semantic misalignment between different modalities leads to gradient conflicts; (2) the message-passing mechanism of graph neural networks further structurally amplifies such gradient noise. Based on this, we propose Structural Regularized Gradient Matching (SR-GM), a condensation framework for multimodal graphs. This method alleviates gradient conflicts between modalities through a gradient decoupling mechanism and introduces a structural damping regularizer to suppress the propagation of gradient noise in the topology, thereby transforming the graph structure from a noise amplifier into a training stabilizer. Extensive experiments on four multimodal graph datasets demonstrate the effectiveness of SR-GM, highlighting its state-of-the-art performance and cross-architecture generalization capabilities in multimodal graph dataset condensation.
format Preprint
id arxiv_https___arxiv_org_abs_2511_20222
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Decoupling and Damping: Structurally-Regularized Gradient Matching for Multimodal Graph Condensation
Shen, Lian
Chen, Zhendan
Song, Meijia
jiang, Yinhui
Su, Ziming
Liu, Juan
Liu, Xiangrong
Machine Learning
In multimodal graph learning, graph structures that integrate information from multiple sources, such as vision and text, can more comprehensively model complex entity relationships. However, the continuous growth of their data scale poses a significant computational bottleneck for training. Graph condensation methods provide a feasible path forward by synthesizing compact and representative datasets. Nevertheless, existing condensation approaches generally suffer from performance limitations in multimodal scenarios, mainly due to two reasons: (1) semantic misalignment between different modalities leads to gradient conflicts; (2) the message-passing mechanism of graph neural networks further structurally amplifies such gradient noise. Based on this, we propose Structural Regularized Gradient Matching (SR-GM), a condensation framework for multimodal graphs. This method alleviates gradient conflicts between modalities through a gradient decoupling mechanism and introduces a structural damping regularizer to suppress the propagation of gradient noise in the topology, thereby transforming the graph structure from a noise amplifier into a training stabilizer. Extensive experiments on four multimodal graph datasets demonstrate the effectiveness of SR-GM, highlighting its state-of-the-art performance and cross-architecture generalization capabilities in multimodal graph dataset condensation.
title Decoupling and Damping: Structurally-Regularized Gradient Matching for Multimodal Graph Condensation
topic Machine Learning
url https://arxiv.org/abs/2511.20222