{"id":"arxiv/2601.10708","url":"https://arxiv.org/abs/2601.10708","tags":["cs.LG","math.ST"],"score":0,"title":"High-accuracy and dimension-free sampling with diffusions","domain":"cs.LG","source":"arxiv","authors":["Khashayar Gatmiry","Sitan Chen","Adil Salim"],"reviews":0,"subject":"cs.LG","summary":"Diffusion models have shown remarkable empirical success in sampling from rich multi-modal distributions. Their inference relies on numerically solving a certain differential equation. This differential equation cannot be solved in closed form, and its resolution via discretization typically requires many small iterations to produce \\emph{high-quality} samples. More precisely, prior works have shown that the iteration complexity of discretization methods for diffusion models scales polynomially in the ambient dimension and the inverse accuracy $1/\\varepsilon$. In this work, we propose a new…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-01-15","reactions":{}}