Research map / Image generation / Fast diffusion sampling
Measurement Score-Based Diffusion Model
Chicago Y. Park, Shirin Shoushtari, Hongyu An, Ulugbek S. Kamilov
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 changed | Where | Measured | Effect |
|---|---|---|---|
| train on partial measurement scores instead of Ambient diffusion (masked data, no noise) | MSM · FFHQ unconditional | FID: 61.46 -> 32.54 | Improved |
| train on noisy+subsampled data with MSM instead of GSURE diffusion | MSM · FFHQ unconditional | FID: 94.88 -> 48.70 | Improved |
| use MSM instead of Ambient diffusion on multi-coil MRI | MSM · fastMRI unconditional R=4 | FID: 47.80 -> 43.60 | Improved |
| increase acceleration to R=8 (ablation) | MSM · fastMRI unconditional | FID: 43.60 -> 74.92 | Hurt |
| add measurement noise eta=0.1 to training data (ablation) | MSM · fastMRI unconditional R=4 | FID: 43.60 -> 79.78 | Hurt |
| use MSM posterior sampling in place of A-DPS | MSM · Box inpainting | PSNR: 20.14 -> 24.71 | Improved |
| use MSM posterior sampling in place of A-DPS | MSM · 4x super-resolution | PSNR: 22.61 -> 28.11 | Improved |
| use MSM instead of A-DPS | MSM · CS-MRI x4 | PSNR: 27.28 -> 30.71 | Improved |
| use MSM instead of Robust SSDU | MSM · CS-MRI x4 | SSIM: 0.847 -> 0.839; PSNR 29.65 -> 30.71 | Mixed |
| use MSM instead of Robust SSDU | MSM · CS-MRI x6 | PSNR: 28.02 -> 28.86 | Improved |
| use A-DPS (Ambient prior) vs degraded input | A-DPS · Box inpainting | PSNR: input 18.26 vs A-DPS 20.14 (SSIM 0.749 vs 0.621) | Hurt |
Closest papers by meaning
- SITCOM: Step-wise Triple-Consistent Diffusion Sampling For Inverse Problems (ICML 2025 · similarity 0.90)
- Provably Robust Score-Based Diffusion Posterior Sampling for Plug-and-Play Image Reconstruction (NeurIPS 2024 · similarity 0.90)
- Learning Single Index Models with Diffusion Priors (ICML 2025 · similarity 0.90)
- A Variational Perspective on Solving Inverse Problems with Diffusion Models (ICLR 2024 · similarity 0.90)
- Ambient Diffusion Posterior Sampling: Solving Inverse Problems with Diffusion Models Trained on Corrupted Data (ICLR 2025 · similarity 0.90)
- Whitened Score Diffusion: A Structured Prior for Imaging Inverse Problems (NeurIPS 2025 · similarity 0.89)
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.
