{"id":"arxiv/2610.02663","url":"https://arxiv.org/abs/2610.02663","tags":["stat.ML","cs.AI","cs.LG","math.ST"],"score":0,"title":"Generalization Bounds for Flow-matching Generative Models for Intrinsically Low-dimensional Data","domain":"stat.ML","source":"arxiv","authors":["Saptarshi Chakraborty","Quentin Berthet","Peter L. Bartlett"],"reviews":0,"subject":"stat.ML","summary":"Despite the remarkable empirical success of flow-matching models, their statistical generalization guarantees remain underdeveloped. Existing analyses often impose restrictive assumptions on the estimated velocity field and yield convergence rates that fail to reflect the intrinsic low-dimensional structure common in real data, such as natural images and molecular geometries. In this work, we study the statistical generalization of flow-matching models for learning an unknown distribution $P_{\\mathrm{data}}$ from finitely many samples. We derive finite-sample error bounds on the learned…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-10-02","reactions":{}}