AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Fast Sampling of Diffusion Models via Operator Learning

Hongkai Zheng, Weili Nie, Arash Vahdat, Kamyar Azizzadenesheli, Anima Anandkumar

ICML 2023 · Image generation › Fast diffusion sampling · approach: Fast diffusion ODE solvers · 17 citations (OpenAlex)

Read the paper

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 DSNO instead of one-evaluation progressive distillationCIFAR-10 diffusion sampling · CIFAR-10FID: 9.12 -> 3.78Improved
Use DSNO instead of one-evaluation progressive distillationImageNet-64 diffusion sampling · ImageNet-64FID: 15.99 -> 7.83Improved
Add temporal convolution blocks to the U-Net (ablation)DSNO on CIFAR-10 · CIFAR-10FID: 8.09 -> 4.23; 7.85 -> 4.12Improved
Use SNR^0.5 loss weighting instead of uniform weighting (ablation)DSNO on CIFAR-10 · CIFAR-10FID: 4.56 -> 4.21Improved
Use quadratic instead of uniform time discretization (ablation)DSNO on CIFAR-10 · CIFAR-10FID: 4.33 -> 4.21Improved
Increase temporal resolution from 2 to 4 (ablation)DSNO on CIFAR-10 · CIFAR-10FID: 5.01 -> 4.21Improved
Increase temporal resolution from 4 to 8 (ablation)DSNO on CIFAR-10 · CIFAR-10FID: 4.21 -> 3.98Improved
Use LPIPS instead of L1 loss (ablation)DSNO on CIFAR-10 · CIFAR-10FID: 4.12 -> 3.78Improved

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