{"id":"arxiv/2609.34613","url":"https://arxiv.org/abs/2609.34613","tags":["stat.ML","cs.LG","math.PR"],"score":0,"title":"Probabilistic Geodesic Flow Matching on Location-Scale Families","domain":"stat.ML","source":"arxiv","authors":["Zeyuan Yu","Zhi Chang","Shiwei Lan"],"reviews":0,"subject":"stat.ML","summary":"Flow matching (FM) has recently emerged as a promising framework for generative modeling due to its conceptual simplicity and strong empirical performance. In FM, samples are transported along a vector field parameterized by a neural network, inducing a probability path that evolves from a simple noise distribution to the target data distribution, governed by an ordinary differential equation (ODE). However, existing FM approaches predominantly rely on probability paths derived from optimal transport (OT) between Gaussian distributions, which may be suboptimal for capturing complex data with…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-09-28","reactions":{}}