SBGD: Improving Graph Diffusion Generative Model via Stochastic Block Diffusion

2025-08-19

AI Summary

Graph diffusion generative models (GDGMs) suffer from poor scalability—due to memory-intensive full-graph storage—and weak size generalization—struggling to generate graphs of unseen scales. To address these limitations, we propose BlockDiff, the first block-based graph diffusion framework that explicitly incorporates structural priors into the diffusion process. BlockDiff projects graphs into a low-dimensional, structure-aware block space via randomized block mapping and performs efficient diffusion modeling therein, thereby avoiding costly full-graph operations. This design reduces memory consumption by up to 6× and inherently supports cross-size generation. Experiments demonstrate that BlockDiff achieves generation quality on par with or surpassing state-of-the-art methods across multiple benchmarks. Moreover, it exhibits significantly improved generalization in zero-shot size-transfer tasks—generating graphs of novel scales without retraining.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Computer Vision: Diffusion Models for Vision
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Web Mining and Content Analysis: Content-based information diffusion

Abstract

Graph diffusion generative models (GDGMs) have emerged as powerful tools for generating high-quality graphs. However, their broader adoption faces challenges in emph{scalability and size generalization}. GDGMs struggle to scale to large graphs due to their high memory requirements, as they typically operate in the full graph space, requiring the entire graph to be stored in memory during training and inference. This constraint limits their feasibility for large-scale real-world graphs. GDGMs also exhibit poor size generalization, with limited ability to generate graphs of sizes different from those in the training data, restricting their adaptability across diverse applications. To address these challenges, we propose the stochastic block graph diffusion (SBGD) model, which refines graph representations into a block graph space. This space incorporates structural priors based on real-world graph patterns, significantly reducing memory complexity and enabling scalability to large graphs. The block representation also improves size generalization by capturing fundamental graph structures. Empirical results show that SBGD achieves significant memory improvements (up to 6$ imes$) while maintaining comparable or even superior graph generation performance relative to state-of-the-art methods. Furthermore, experiments demonstrate that SBGD better generalizes to unseen graph sizes. The significance of SBGD extends beyond being a scalable and effective GDGM; it also exemplifies the principle of modularization in generative modeling, offering a new avenue for exploring generative models by decomposing complex tasks into more manageable components.

Problem

• Addressing scalability issues in graph diffusion models for large graphs
• Improving size generalization to generate graphs of unseen sizes
• Reducing memory complexity while maintaining generation performance

Innovation

• Stochastic block diffusion refines graph representations
• Reduces memory complexity with structural priors
• Improves size generalization via block graph space