Towards Fast Graph Generation via Autoregressive Noisy Filtration Modeling

2025-02-04

AI Summary

Graph generation models have long struggled to simultaneously achieve high-fidelity modeling of complex graph distributions and computational efficiency. This paper proposes a filtration-based autoregressive graph generation framework, which encodes graphs as monotonically increasing subgraph sequences—introducing the topological data analysis concept of *filtration* to graph generation for the first time and enabling compact sequence modeling. Methodologically, we design an autoregressive graph mixer, a noise-augmentation mechanism, and a reinforcement learning strategy to systematically identify and mitigate exposure bias. Evaluated on diverse synthetic and real-world graph benchmarks, our approach matches state-of-the-art diffusion models in generation quality while reducing sequence length by two orders of magnitude and accelerating inference by 100×, thereby substantially breaking the traditional quality–efficiency trade-off.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Natural Language Processing: Generation
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Graph generative models often face a critical trade-off between learning complex distributions and achieving fast generation speed. We introduce Autoregressive Noisy Filtration Modeling (ANFM), a novel approach that addresses both challenges. ANFM leverages filtration, a concept from topological data analysis, to transform graphs into short sequences of monotonically increasing subgraphs. This formulation extends the sequence families used in previous autoregressive models. To learn from these sequences, we propose a novel autoregressive graph mixer model. Our experiments suggest that exposure bias might represent a substantial hurdle in autoregressive graph generation and we introduce two mitigation strategies to address it: noise augmentation and a reinforcement learning approach. Incorporating these techniques leads to substantial performance gains, making ANFM competitive with state-of-the-art diffusion models across diverse synthetic and real-world datasets. Notably, ANFM produces remarkably short sequences, achieving a 100-fold speedup in generation time compared to diffusion models. This work marks a significant step toward high-throughput graph generation.

Problem

• Improve graph generation speed
• Reduce exposure bias impact
• Enhance complex distribution learning

Innovation

• Autoregressive Noisy Filtration Modeling
• Sequence-based graph transformation
• Noise augmentation and reinforcement learning