用拓扑过滤思想加速图生成，速度超扩散模型100倍且质量不降。
Fast Graph Generation via Autoregressive Noisy Filtration Modeling
- 将图转化为子图序列，用自回归建模非单调生成过程。
- 在多个数据集上达到与扩散模型相当的生成质量，推理速度快100倍以上。
- 适合需要高吞吐图生成的场景，如药物分子设计或网络模拟。
现有图生成模型常在样本质量和生成速度间权衡。本文提出自回归噪声过滤建模（ANFM），一种灵活的自回归框架，同时解决这两类挑战。ANFM利用拓扑数据分析中的过滤概念，将图转换为短序列的子图。我们识别出自回归图生成中的暴露偏差问题，并提出噪声增强与强化学习作为有效缓解策略，使ANFM能学习边的添加与删除操作。这一独特能力允许其在生成过程中纠正错误，通过建模非单调图序列实现。实验结果表明，ANFM在质量上媲美最先进扩散模型，同时推理速度提升超过100倍，是高吞吐图生成的有力方案。源代码已公开于 https://github.com/BorgwardtLab/anfm。
原文摘要 · Abstract (English)
Existing graph generative models often face a critical trade-off between sample quality and generation speed. We introduce Autoregressive Noisy Filtration Modeling (ANFM), a flexible autoregressive framework that addresses both challenges. ANFM leverages filtration, a concept from topological data analysis, to transform graphs into short sequences of subgraphs. We identify exposure bias as a potential hurdle in autoregressive graph generation and propose noise augmentation and reinforcement learning as effective mitigation strategies, which allow ANFM to learn both edge addition and deletion operations. This unique capability enables ANFM to correct errors during generation by modeling non-monotonic graph sequences. Our results show that ANFM matches state-of-the-art diffusion models in quality while offering over 100 times faster inference, making it a promising approach for high-throughput graph generation. The source code is publicly available at https://github.com/BorgwardtLab/anfm .
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。