AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

DMQ: Dissecting Outliers of Diffusion Models for Post-Training Quantization

Dongyeun Lee, Jiwan Hur, Hyounguk Shon, Jae Young Lee, Junmo Kim

ICCV 2025 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 3 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
replace SmoothQuant scaling with learned equivalent scalingW4A8 quantized LDM-4 · FFHQ 256x256FID: 454.16 -> 30.37Improved
add Learned Equivalent Scaling (ablation)baseline PTQ · FFHQ W4A8FID: 36.08 -> 33.46Improved
add adaptive timestep weighting (ablation)LES · FFHQ W4A8FID: 33.46 -> 31.83Improved
add Power-of-Two Scaling (ablation)LES + weighting · FFHQ W4A8FID: 31.83 -> 30.37Improved
add Power-of-Two Scaling (ablation)LES + weighting · FFHQ W4A8sFID: 24.39 -> 22.72Improved
use adaptive weighting instead of uniform (ablation)LES · FFHQ W4A8FID: 33.46 -> 31.83Improved
use linear early-step weighting instead of uniform (ablation)LES · FFHQ W4A8FID: 33.46 -> 36.58Hurt
use quadratic early-step weighting instead of uniform (ablation)LES · FFHQ W4A8FID: 33.46 -> 35.30Hurt
select PTS factors by voting instead of MSE (ablation)PTS on all layers · FFHQ W4A8FID: 32.55 -> 31.91Improved
apply PTS only to skip connections instead of all layers (ablation)voting PTS · FFHQ W4A8FID: 31.91 -> 30.37Improved
add MSE-selected PTS to all layers (ablation)w/o PTS · FFHQ W4A8FID: 31.83 -> 32.55Hurt
custom W4A8 kernel with bit-shifting vs FP32 GEMMPTS · latency at M=3072speedup: 5.17xImproved

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