#machine learning
Jan 2026
Variational Bayesian Flow Network for Graph Generation
Variational Bayesian Flow Network is proposed, which performs a variational lifting to a tractable joint Gaussian variational belief family governed by structured precisions, and improves fidelity and diversity, and surpasses baseline methods on synthetic and molecular graph datasets.
Yi-Da Xiong, Jia-Meng Chen, Xiuwen Gong et al.
· arXiv.org · 1 citation