Id: arxiv/2611.09980 Url: https://arxiv.org/abs/2611.09980 Tags: cs.CV, math.NA, math.OC Score: 0 Title: Preconditioned Implicit Midpoint Langevin Sampling for Non-Smooth Bayesian Imaging Domain: cs.CV Source: arxiv Authors: Sichen Wang, Zhipeng Lu Reviews: 0 Subject: cs.CV Category: research Comments: 0 Disputed: false Division: cs Mentions: 0 Discussed: 0 Published: 2026-11-01 Page: https://delphi.science/r/arxiv/2611.09980 We introduce P-IMLA, a preconditioned implicit-midpoint Langevin sampler for high-dimensional log-concave targets with non-smooth regularization. The obstruction is that the large steps enabled by matrix preconditioning are precisely where preconditioned Euler-Maruyama inflates stationary variance. Implicit midpoint removes this surplus via a Cayley-transform identity, making P-IMLA Gaussian-exact at any step size. Beyond Gaussians, whitening transfers the sharp IMLA contraction rate to $κ_{\mathrm{eff}}$, and backward-error analysis gives drift-only rather than Euler-type diffusion bias…