AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Measurement Score-Based Diffusion Model

Chicago Y. Park, Shirin Shoushtari, Hongyu An, Ulugbek S. Kamilov

ICLR 2026 · Image generation › Fast diffusion sampling · approach: Diffusion theory and design space · 8 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
train on partial measurement scores instead of Ambient diffusion (masked data, no noise)MSM · FFHQ unconditionalFID: 61.46 -> 32.54Improved
train on noisy+subsampled data with MSM instead of GSURE diffusionMSM · FFHQ unconditionalFID: 94.88 -> 48.70Improved
use MSM instead of Ambient diffusion on multi-coil MRIMSM · fastMRI unconditional R=4FID: 47.80 -> 43.60Improved
increase acceleration to R=8 (ablation)MSM · fastMRI unconditionalFID: 43.60 -> 74.92Hurt
add measurement noise eta=0.1 to training data (ablation)MSM · fastMRI unconditional R=4FID: 43.60 -> 79.78Hurt
use MSM posterior sampling in place of A-DPSMSM · Box inpaintingPSNR: 20.14 -> 24.71Improved
use MSM posterior sampling in place of A-DPSMSM · 4x super-resolutionPSNR: 22.61 -> 28.11Improved
use MSM instead of A-DPSMSM · CS-MRI x4PSNR: 27.28 -> 30.71Improved
use MSM instead of Robust SSDUMSM · CS-MRI x4SSIM: 0.847 -> 0.839; PSNR 29.65 -> 30.71Mixed
use MSM instead of Robust SSDUMSM · CS-MRI x6PSNR: 28.02 -> 28.86Improved
use A-DPS (Ambient prior) vs degraded inputA-DPS · Box inpaintingPSNR: input 18.26 vs A-DPS 20.14 (SSIM 0.749 vs 0.621)Hurt

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