Variational Bayesian Flow Network for Graph Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xiong, Yida, Chen, Jiameng, Gong, Xiuwen, Wu, Jia, Pan, Shirui, Hu, Wenbin
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912861984391168
author Xiong, Yida
Chen, Jiameng
Gong, Xiuwen
Wu, Jia
Pan, Shirui
Hu, Wenbin
author_facet Xiong, Yida
Chen, Jiameng
Gong, Xiuwen
Wu, Jia
Pan, Shirui
Hu, Wenbin
contents Graph generation aims to sample discrete node and edge attributes while satisfying coupled structural constraints. Diffusion models for graphs often adopt largely factorized forward-noising, and many flow-matching methods start from factorized reference noise and coordinate-wise interpolation, so node-edge coupling is not encoded by the generative geometry and must be recovered implicitly by the core network, which can be brittle after discrete decoding. Bayesian Flow Networks (BFNs) evolve distribution parameters and naturally support discrete generation. But classical BFNs typically rely on factorized beliefs and independent channels, which limit geometric evidence fusion. We propose Variational Bayesian Flow Network (VBFN), which performs a variational lifting to a tractable joint Gaussian variational belief family governed by structured precisions. Each Bayesian update reduces to solving a symmetric positive definite linear system, enabling coupled node and edge updates within a single fusion step. We construct sample-agnostic sparse precisions from a representation-induced dependency graph, thereby avoiding label leakage while enforcing node-edge consistency. On synthetic and molecular graph datasets, VBFN improves fidelity and diversity, and surpasses baseline methods.
format Preprint
id arxiv_https___arxiv_org_abs_2601_22524
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Variational Bayesian Flow Network for Graph Generation
Xiong, Yida
Chen, Jiameng
Gong, Xiuwen
Wu, Jia
Pan, Shirui
Hu, Wenbin
Machine Learning
Graph generation aims to sample discrete node and edge attributes while satisfying coupled structural constraints. Diffusion models for graphs often adopt largely factorized forward-noising, and many flow-matching methods start from factorized reference noise and coordinate-wise interpolation, so node-edge coupling is not encoded by the generative geometry and must be recovered implicitly by the core network, which can be brittle after discrete decoding. Bayesian Flow Networks (BFNs) evolve distribution parameters and naturally support discrete generation. But classical BFNs typically rely on factorized beliefs and independent channels, which limit geometric evidence fusion. We propose Variational Bayesian Flow Network (VBFN), which performs a variational lifting to a tractable joint Gaussian variational belief family governed by structured precisions. Each Bayesian update reduces to solving a symmetric positive definite linear system, enabling coupled node and edge updates within a single fusion step. We construct sample-agnostic sparse precisions from a representation-induced dependency graph, thereby avoiding label leakage while enforcing node-edge consistency. On synthetic and molecular graph datasets, VBFN improves fidelity and diversity, and surpasses baseline methods.
title Variational Bayesian Flow Network for Graph Generation
topic Machine Learning
url https://arxiv.org/abs/2601.22524