用更灵活的距离函数改进生成模型更新机制，提升生成质量。
General Proximal Flow Networks
- 用任意散度替代固定KL散度，统一生成建模的迭代更新框架。
- 在多个数据集上验证，适配数据几何的散度能显著提升生成质量。
- 适合研究生成模型优化与变分推断的学者，尤其关注可扩展框架者。
本文提出广义近端流网络（GPFNs），作为贝叶斯流网络的推广，拓展了可接受信念更新算子的类别。在贝叶斯流网络中，每步更新是基于KL散度的贝叶斯后验更新，等价于关于KL散度的近端步骤。GPFNs将这一固定选择替换为任意散度或距离函数（如Wasserstein距离），建立统一的近端算子框架用于迭代生成建模。推导出相应的训练与采样流程，建立起与近端优化的正式联系，并将标准BFN更新作为特例恢复。实验表明，根据数据几何自适应选择散度可显著提升生成质量，凸显该框架的实用性。
原文摘要 · Abstract (English)
This paper introduces General Proximal Flow Networks (GPFNs), a generalization of Bayesian Flow Networks that broadens the class of admissible belief-update operators. In Bayesian Flow Networks, each update step is a Bayesian posterior update, which is equivalent to a proximal step with respect to the Kullback-Leibler divergence. GPFNs replace this fixed choice with an arbitrary divergence or distance function, such as the Wasserstein distance, yielding a unified proximal-operator framework for iterative generative modeling. The corresponding training and sampling procedures are derived, establishing a formal link to proximal optimization and recovering the standard BFN update as a special case. Empirical evaluations confirm that adapting the divergence to the underlying data geometry yields measurable improvements in generation quality, highlighting the practical benefits of this broader framework.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。