AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Adversarial Distribution Matching for Diffusion Distillation Towards Efficient Image and Video Synthesis

Yanzuo Lu, Yuxi Ren, Xin Xia, Shanchuan Lin, Xing Wang, Xuefeng Xiao, Andy J. Ma, Xiaohua Xie, Jian-Huang Lai

ICCV 2025 · Image generation › Fast diffusion sampling · approach: Distillation for few-step generation · 20 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
add ADM fine-tuning after ADP pre-training (ablation)DMDX one-step SDXL · COCO 10K (SDXL ablation)HPSv2: 27.3499 -> 27.7046Improved
add ADM fine-tuning after ADP pre-training (ablation)DMDX one-step SDXL · COCO 10K (SDXL ablation)MPS: 10.6646 -> 11.1978Improved
add ADP pre-training before ADM (ablation)ADM one-step SDXL · COCO 10K (SDXL ablation)HPSv2: 26.8732 -> 27.7046Improved
replace DMD loss with ADM (both after ADP) (ablation)DMDX one-step SDXL · COCO 10K (SDXL ablation)HPSv2: 27.3985 -> 27.7046Improved
replace DMD loss with ADM (both without ADP) (ablation)one-step SDXL w/o ADP · COCO 10K (SDXL ablation)HPSv2: 25.9680 -> 26.8732Improved
use SAM instead of DINOv2 as pixel-space discriminator (ablation)ADP pre-training · COCO 10K (SDXL ablation)HPSv2: 27.1039 -> 27.3499Improved
add pixel-space discriminator weight lambda2=0.15 (vs 0.0) (ablation)ADP pre-training · COCO 10K (SDXL ablation)HPSv2: 26.7869 -> 27.3499Improved
lower pixel-space weight from 0.3 to 0.15 (ablation)ADP pre-training · COCO 10K (SDXL ablation)HPSv2: 26.8902 -> 27.3499Improved
ADP adversarial pre-training instead of single Rectified Flow reflow (ablation)one-step SDXL pre-training · COCO 10K (SDXL ablation)HPSv2: 23.6093 -> 27.3499Improved
increase TTUR from 1 to 8 (ablation)DMDX one-step SDXL · COCO 10KHPSv2: 27.7046 -> 27.7586 with training time x1.00 -> x2.53No clear effect
train 2x longer (ablation)ADM 8-step CogVideoX-2b · VBenchfinal score: 78.584 -> 80.764Improved
add CFG to the few-step generator (ablation)ADM 8-step CogVideoX-2b · VBenchsemantic score: 69.621 -> 75.569Improved

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