AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Metropolis-Adjusted Diffusion Models

Kevin H. Lam, Tyler Farghly, Christopher Williams, Jun Yang, Yee Whye Teh, Arnaud Doucet

NeurIPS 2026 · Image generation › Fast diffusion sampling · approach: Diffusion theory and design space · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes Metropolis-adjusted Langevin correctors for diffusion models (MADM) using only the score, via an exact two-coin Bernoulli factory Barker correction and a Simpson's-rule MH approximation, showing gains in FID on CIFAR-10, FFHQ, AFHQv2 and ImageNet-64.

Problem

placeholder

Approach

placeholder

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add one Simpson-rule MADM corrector step per predictor stepPF-ODE with Heun solver · CIFAR-10FID: 1.96 -> 1.95Improved
add one Simpson-rule MADM corrector step per predictor stepPF-ODE with Heun solver · FFHQFID: 2.47 -> 2.43Improved
add one Simpson-rule MADM corrector step per predictor stepPF-ODE with Heun solver · AFHQv2FID: 2.04 -> 2.03Improved
add one Simpson-rule MADM corrector step per predictor stepPF-ODE with Heun solver · ImageNet-64FID: 2.32 -> 2.15Improved
add one Simpson-rule MADM corrector step per predictor stepPF-ODE with Euler solver · CIFAR-10FID: 7.62 -> 7.52Improved
add one Simpson-rule MADM corrector step per predictor stepPF-ODE with Euler solver · FFHQFID: 4.59 -> 4.52Improved
add one Simpson-rule MADM corrector step per predictor stepPF-ODE with Euler solver · AFHQv2FID: 2.93 -> 2.91Improved
add one Simpson-rule MADM corrector step per predictor stepPF-ODE with Euler solver · ImageNet-64FID: 2.41 -> 2.25Improved
Euler predictor with MADM instead of higher-order Heun predictor without correctorPF-ODE · ImageNet-64FID: 2.32 -> 2.25Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy