aiqbench analysis
This paper introduces the Variational Bayesian Flow Network (VBFN), a novel architecture designed to address node-edge coupling limitations in existing graph generation models. By leveraging structured precisions within a joint Gaussian variational belief family, VBFN enables simultaneous updates for nodes and edges, significantly improving fidelity and diversity on synthetic and molecular datasets compared to baseline methods. The approach offers a robust alternative to factorized diffusion models for tasks requiring strict structural constraints in discrete graph sampling.
Variational Bayesian Flow Network for Graph Generation
arXiv:2601.22524v2 Announce Type: replace Abstract: 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. The reproducible code is available at https://github.com/Cello2195/VBFN.
Abstract: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. The reproducible code is available at this https URL.
Machine Learning (cs.LG)
arXiv:2601.22524 [cs.LG]
(or arXiv:2601.22524v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2601.22524 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 271 words
Related coverage
Joint Bayesian Inference of Graphical Structure and Parameters with a Single Generative Flow Network
arXiv · Aug 31, 2026
High-Resolution Dynamic Functional Connectivity Generation with Graph-Variate Flow Matching
arXiv · Sep 30, 2026
Permutation-Equivariant Flow Matching for Alignment-Free Neural Weight Generation
arXiv · Sep 30, 2026
Navigating uncertainty in Amazon's middle-mile network
Amazon Science · May 6, 2026