{"id":"arxiv/2602.22369","url":"https://arxiv.org/abs/2602.22369","tags":["math.ST","math.PR","stat.ML"],"score":0,"title":"Sampling from Constrained Gibbs Measures: with Applications to High-Dimensional Bayesian Inference","domain":"math.ST","source":"arxiv","authors":["Ruixiao Wang","Xiaohong Chen","Sinho Chewi"],"reviews":0,"subject":"math.ST","summary":"This paper considers a non-standard problem of generating samples from a low-temperature Gibbs distribution with \\emph{constrained} support, when some of the coordinates of the mode lie on the boundary. These coordinates are referred to as the non-regular part of the model. We show that in a ``pre-asymptotic'' regime in which the limiting Laplace approximation is not yet valid, the low-temperature Gibbs distribution concentrates on a neighborhood of its mode. Within this region, the distribution is a bounded perturbation of a product measure: a strongly log-concave distribution in the…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-02-25","reactions":{}}