arXiv：cs.AI· Chenrui Ma, Xi Xiao, Lin Zhao, Xinru Jiang, Tianyang Wang, Yanning Shen, Ferdinando Fioretto·· 6 小时前
Posterior Flow Matching：让流匹配同时输出端点不确定性
Posterior Flow Matching
AI 导读
研究者提出 Posterior Flow Matching（PFM），一种用高斯后验模型学习条件速度均值与逐分量方差的生成建模框架，单次模型评估即可同时给出传输预测和端点不确定性。该框架支持均值驱动的 ODE 与后验驱动的 SDE，后者局部噪声尺度会随预测不确定性自适应调整。在图像、点云、分子和动作生成任务上，PFM 提升了生成质量，并能在整个生成过程中提供端点后验读数。
正文
Abstract:Flow matching learns how to transport noise toward data, but its mean prediction does not directly report the endpoint uncertainty remaining at an intermediate state. This paper introduces Posterior Flow Matching (PFM), a general generative modeling framework that learns conditional velocity means and componentwise variances with a Gaussian posterior model. The same model evaluation supplies both a transport prediction and endpoint uncertainty, making posterior diagnostics available throughout generation. These learned moments support a mean-driven ordinary differential equation (ODE) and a posterior-driven stochastic differential equation (SDE), whose local noise scales adapt to the predicted uncertainty. Across image, point-cloud, molecular, and action generation, PFM improves generation quality and provides endpoint-posterior readouts throughout generation.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2610.13994 [cs.LG] |
| (or arXiv:2610.13994v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.13994 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Chenrui Ma [view email]
[v1]
Sat, 26 Sep 2026 20:44:52 UTC (30,029 KB)
来源：arXiv：cs.AI · arxiv.org