{"id":"arxiv/2602.10989","url":"https://arxiv.org/abs/2602.10989","tags":["math.ST","cs.IT","cs.LG","math.PR","stat.ML"],"score":0,"title":"Variational Optimality of Föllmer Processes in Generative Diffusions","domain":"math.ST","source":"arxiv","authors":["Yifan Chen","Eric Vanden-Eijnden"],"reviews":0,"subject":"math.ST","summary":"We construct and analyze generative diffusions that transport a point mass to a prescribed target distribution over a finite time horizon using the stochastic interpolant framework. The drift is expressed as a conditional expectation that can be estimated from independent samples without simulating stochastic processes. We show that the diffusion coefficient can be tuned \\emph{a~posteriori} without changing the time-marginal distributions. Among all such tunings, we prove that minimizing the impact of estimation error on the path-space Kullback--Leibler divergence selects, in closed form, a…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-02-11","reactions":{}}