Smooth Interpolation for Improved Discrete Graph Generative Models

Yuxuan Song, Juntong Shi, Jingjing Gong, Minkai Xu, Stefano Ermon, Hao Zhou, Wei-Ying Ma
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:56363-56388, 2025.

Abstract

Though typically represented by the discrete node and edge attributes, the graph topological information can be sufficiently captured by the graph spectrum in a continuous space. It is believed that incorporating the continuity of graph topological information into the generative process design could establish a superior paradigm for graph generative modeling. Motivated by such prior and recent advancements in the generative paradigm, we propose Graph Bayesian Flow Networks (GraphBFN) in this paper, a principled generative framework that designs an alternative generative process emphasizing the dynamics of topological information. Unlike recent discrete-diffusion-based methods, GraphBFNemploys the continuous counts derived from sampling infinite times from a categorical distribution as latent to facilitate a smooth decomposition of topological information, demonstrating enhanced effectiveness. To effectively realize the concept, we further develop an advanced sampling strategy and new time-scheduling techniques to overcome practical barriers and boost performance. Through extensive experimental validation on both generic graph and molecular graph generation tasks, GraphBFN could consistently achieve superior or competitive performance with significantly higher training and sampling efficiency.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-song25f, title = {Smooth Interpolation for Improved Discrete Graph Generative Models}, author = {Song, Yuxuan and Shi, Juntong and Gong, Jingjing and Xu, Minkai and Ermon, Stefano and Zhou, Hao and Ma, Wei-Ying}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {56363--56388}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/song25f/song25f.pdf}, url = {https://proceedings.mlr.press/v267/song25f.html}, abstract = {Though typically represented by the discrete node and edge attributes, the graph topological information can be sufficiently captured by the graph spectrum in a continuous space. It is believed that incorporating the continuity of graph topological information into the generative process design could establish a superior paradigm for graph generative modeling. Motivated by such prior and recent advancements in the generative paradigm, we propose Graph Bayesian Flow Networks (GraphBFN) in this paper, a principled generative framework that designs an alternative generative process emphasizing the dynamics of topological information. Unlike recent discrete-diffusion-based methods, GraphBFNemploys the continuous counts derived from sampling infinite times from a categorical distribution as latent to facilitate a smooth decomposition of topological information, demonstrating enhanced effectiveness. To effectively realize the concept, we further develop an advanced sampling strategy and new time-scheduling techniques to overcome practical barriers and boost performance. Through extensive experimental validation on both generic graph and molecular graph generation tasks, GraphBFN could consistently achieve superior or competitive performance with significantly higher training and sampling efficiency.} }
Endnote
%0 Conference Paper %T Smooth Interpolation for Improved Discrete Graph Generative Models %A Yuxuan Song %A Juntong Shi %A Jingjing Gong %A Minkai Xu %A Stefano Ermon %A Hao Zhou %A Wei-Ying Ma %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-song25f %I PMLR %P 56363--56388 %U https://proceedings.mlr.press/v267/song25f.html %V 267 %X Though typically represented by the discrete node and edge attributes, the graph topological information can be sufficiently captured by the graph spectrum in a continuous space. It is believed that incorporating the continuity of graph topological information into the generative process design could establish a superior paradigm for graph generative modeling. Motivated by such prior and recent advancements in the generative paradigm, we propose Graph Bayesian Flow Networks (GraphBFN) in this paper, a principled generative framework that designs an alternative generative process emphasizing the dynamics of topological information. Unlike recent discrete-diffusion-based methods, GraphBFNemploys the continuous counts derived from sampling infinite times from a categorical distribution as latent to facilitate a smooth decomposition of topological information, demonstrating enhanced effectiveness. To effectively realize the concept, we further develop an advanced sampling strategy and new time-scheduling techniques to overcome practical barriers and boost performance. Through extensive experimental validation on both generic graph and molecular graph generation tasks, GraphBFN could consistently achieve superior or competitive performance with significantly higher training and sampling efficiency.
APA
Song, Y., Shi, J., Gong, J., Xu, M., Ermon, S., Zhou, H. & Ma, W.. (2025). Smooth Interpolation for Improved Discrete Graph Generative Models. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:56363-56388 Available from https://proceedings.mlr.press/v267/song25f.html.

Related Material