AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Efficient Diffusion Training via Min-SNR Weighting Strategy

Tiankai Hang, Shuyang Gu, Chen Li, Jianmin Bao, Dong Chen, Han Hu, Xin Geng, Baining Guo

ICCV 2023 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 189 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
weight diffusion loss by min(SNR, gamma) with gamma=5ViT-B on ImageNet 256x256 · ImageNet 256x256convergence speed to FID 10: 3.4x speedupImproved
add Min-SNR-5 weighting when predicting x0UNet · ImageNet 256x256, 200K iterationsFID: 25.93 -> 7.99Improved
add Min-SNR-5 weighting when predicting epsilonUNet · ImageNet 256x256, 1M iterationsFID: 4.21 -> 4.14Improved
use plain SNR weighting instead of Min-SNR (ablation)ViT-B diffusion training · ImageNet 256x256convergenceHurt
use constant or Max-SNR-gamma weighting when network outputs noise (ablation)diffusion training · ImageNet 256x256training stabilityHurt
increase gamma from 5 to 20 (ablation)ViT-B predicting x0 · ImageNet 256x256FID: 4.92 -> 5.45Hurt
vary gamma on UNet predicting epsilon (ablation)UNet · ImageNet 256x256FID: 4.14 -> 4.12No clear effect
use fixed Min-SNR-gamma weights instead of per-iteration Pareto/UGD optimized weights (ablation)diffusion loss weighting · Pareto objectivePareto objective valueNo clear effect
apply Min-SNR-gamma to the EDM frameworkEDM · ImageNet 64x64convergenceImproved

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