{"id":"arxiv/2602.00869","url":"https://arxiv.org/abs/2602.00869","tags":["cs.LG","cs.AI","math.NA"],"score":0,"title":"Improving Flow Matching by Aligning Flow Divergence","domain":"cs.LG","source":"arxiv","authors":["Yuhao Huang","Taos Transue","Shih-Hsin Wang","William Feldman","Hong Zhang","Bao Wang"],"reviews":0,"subject":"cs.LG","summary":"Conditional flow matching (CFM) stands out as an efficient, simulation-free approach for training flow-based generative models, achieving remarkable performance for data generation. However, CFM is insufficient to ensure accuracy in learning probability paths. In this paper, we introduce a new partial differential equation characterization for the error between the learned and exact probability paths, along with its solution. We show that the total variation gap between the two probability paths is bounded above by a combination of the CFM loss and an associated divergence loss. This…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-01-31","reactions":{}}