{"id":"arxiv/2609.39531","url":"https://arxiv.org/abs/2609.39531","tags":["math.ST","cs.LG"],"score":0,"title":"WEIRDO: WEak resIdual Regularized DOob's h-transform diffusion alignment","domain":"math.ST","source":"arxiv","authors":["Denis Suchkov"],"reviews":0,"subject":"math.ST","summary":"We study the problem of estimating the guidance that steers the distribution learned by a diffusion generative model toward a tilted target $q_0 \\propto w\\,p_0$ at inference time. Relying on the stochastic optimal control approach, we observe that the exact drift correction is the gradient of the logarithm of Doob's $h$-function, and we study the problem of estimating it from a sample. In the present paper, we assume that the score of the pretrained model is available, that the tilting weight is bounded and positive, and that the reference distribution has a bounded support, no smoothness of…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-09-30","reactions":{}}