AI Insight
This paper introduces Variational Bayesian Flow Network (VBFN), a new method for generating graph structures that better preserves the relationships between nodes and edges. Unlike existing diffusion and flow-matching models that treat nodes and edges independently, VBFN uses a joint variational belief system with structured precisions to update coupled node-edge attributes simultaneously through symmetric positive definite linear systems. Experiments on synthetic and molecular graph datasets demonstrate improved fidelity and diversity compared to baseline approaches.
Why it matters
This advancement could improve AI-driven molecular design and drug discovery by generating more realistic graph structures that maintain essential node-edge dependencies. The method's ability to enforce structural consistency has potential applications in materials science, chemical synthesis planning, and any domain requiring accurate generation of complex relational data.
Understand the Science
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
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.
Source: Variational Bayesian Flow Network for Graph Generation