AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Local MAP Sampling for Diffusion Models

Shaorong Zhang, Rob Brekelmans, Greg Ver Steeg

ICML 2026 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Simulation-based inference · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use LMAPS local MAP sampling instead of DAPS posterior samplingDAPS · FFHQ 256 SR 4xPSNR: 29.07 -> 30.74Improved
use LMAPS clean-space local MAP instead of DMAP latent-space local MAPDMAP · FFHQ 256 Gaussian DeblurringPSNR: 26.83 -> 30.88Improved
use LMAPS instead of DMAPDMAP · FFHQ 256 SR 4xFID: 44.78 -> 80.63Hurt
use LMAPS instead of DAPS on random inpaintingDAPS · FFHQ 256 Inpaint (Random)PSNR: 31.12 -> 34.51Improved
use LMAPS instead of DAPS on motion deblurringDAPS · FFHQ 256 Motion DeblurringPSNR: 29.66 -> 32.62Improved
use LMAPS instead of RED-diff on nonlinear deblurringRED-diff · FFHQ 256 Nonlinear DeblurringPSNR: 30.86 -> 29.93Hurt
use LMAPS instead of SITCOM on ImageNet SR 4xSITCOM · ImageNet 256 SR 4xPSNR: 27.07 -> 26.72Hurt
use LMAPS instead of DDNM on ImageNet random inpaintingDDNM · ImageNet 256 Inpaint (Random)PSNR: 31.16 -> 30.59Hurt
use LMAPS on scientific inverse problems versus best baselineRED-diff / DAPS / DDNM · InverseBench LISPSNR: 36.56 -> 38.07 (NR=360)Improved
use LMAPS instead of PnP-DM on CS-MRI 4x noiselessPnP-DM · InverseBench CS-MRIPSNR: 31.80 -> 32.83Improved
increase gradient optimization steps per diffusion step at fixed NFE budget (ablation)LMAPS · Gaussian deblurringreconstruction qualityImproved
use LMAPS in place of SITCOM, which backpropagates through the U-Net with 600 NFEsSITCOM · Gaussian deblurringperformance vs computeNo clear effect

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