{"id":"arxiv/2603.24495","url":"https://arxiv.org/abs/2603.24495","tags":["math.ST","stat.ML"],"score":0,"title":"Reflected diffusion models adapt to low-dimensional data","domain":"math.ST","source":"arxiv","authors":["Asbjørn Holk","Claudia Strauch","Lukas Trottner"],"reviews":0,"subject":"math.ST","summary":"While the mathematical foundations of score-based generative models are increasingly well understood for unconstrained Euclidean spaces, many practical applications involve data restricted to bounded domains. This paper provides a statistical analysis of reflected diffusion models on the hypercube $[0,1]^D$ for target distributions supported on $d$-dimensional linear subspaces. A primary challenge in this setting is the absence of Gaussian transition kernels, which play a central role in standard theory in $\\mathbb{R}^D$. By employing an easily implementable infinite series expansion of the…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-03-25","reactions":{}}