Robust Graph Diffusion Model

ICLR 2026 Conference Submission20370 Authors

Keywords: Diffusion Models, Robust Graph Diffusion Model, Graph Adversarial Attacks

TL;DR: We propose Robust Graph Diffusion Model (RGDM) which performs node-level graph augmentation and generates relatively clean synthetic graph structure for training robust GNNs against various graph adversarial attacks.

Abstract:

Diffusion models represent a powerful class of generative models known for their solid theoretical foundations and remarkable performance across diverse tasks and domains. While diffusion models have been extensively utilized for generating entire graphs or small-scale graphs, no diffusion-based approaches have been developed to synthesize graph structures within an existing graph, including synthetic nodes and their associated edges. In this study, we introduce the Robust Graph Diffusion Model (RGDM), designed to generate labeled synthetic graph structures consisting of nodes and edges that integrate seamlessly into a given graph.
The RGDM consists of a Robust Graph Autoencoder (RGAE) and a Latent Diffusion Model (LDM). Leveraging an edge selection mechanism and an innovative low-rank regularization on the latent feature, the RGDM produces clean and high-quality synthetic graph structures, even when trained on graphs subject to adversarial attacks. Comprehensive experimental evaluations reveal that Graph Neural Networks (GNNs) trained on the augmented graph, which is formed by merging the original attacked graph with the synthetic graph structures, exhibit significantly improved robustness against various graph adversarial attacks in the context of semi-supervised node classification.
The code of the RGDM is available at \url{https://anonymous.4open.science/r/RGDM}.

Primary Area: generative models

Submission Number: 20370