AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation

Zehong Ma, Longhui Wei, Shuai Wang, Shiliang Zhang, Qi Tian

CVPR 2026 · Image generation › Autoregressive and latent generation · approach: High-resolution diffusion synthesis · 39 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 lightweight pixel decoder for high-frequency detail, DiT handles low-frequency semantics (ablation)pixel diffusion baseline (DiT-L) · ImageNet 256x256, 200k iterationsFID: 61.10 -> 34.12Improved
add frequency-aware flow-matching loss (ablation)DeCo · ImageNet 256x256, 200k iterationsFID: 34.12 -> 31.35Improved
add REPA alignment loss (ablation)DeCo · ImageNet 256x256, 200k iterationsFID: 67.55 -> 31.35Improved
use AdaLN interaction instead of upsample-and-add (ablation)DeCo · ImageNet 256x256, 200k iterationsFID: 36.02 -> 31.35Improved
use pixel decoder patch size 1 instead of 16 (ablation)DeCo · ImageNet 256x256, 200k iterationsFID: 55.59 -> 31.35Improved
use pixel decoder patch size 1 instead of 4 (ablation)DeCo · ImageNet 256x256, 200k iterationsFID: 34.39 -> 31.35Improved
increase pixel decoder hidden size from 32 to 64 (ablation)DeCo · ImageNet 256x256, 200k iterationsFID: 34.12 -> 35.88No clear effect
increase pixel decoder depth from 3 to 6 (ablation)DeCo · ImageNet 256x256, 200k iterationsFID: 34.12 -> 35.46Hurt
reduce pixel decoder depth from 3 to 1 (ablation)DeCo · ImageNet 256x256, 200k iterationsFID: 34.12 -> 37.10Hurt
set JPEG quality to 100 (equal frequency weights) in frequency-aware loss (ablation)DeCo · ImageNet 256x256, 200k iterationsFID: 31.35 -> 33.84Hurt
reduce JPEG quality to 50 (ablation)DeCo frequency-aware loss · ImageNet 256x256, 200k iterationsFID: 31.35 -> 31.54No clear effect
change frequency loss weight from 1 to 0.5 or 2 (ablation)DeCo · ImageNet 256x256, 200k iterationsFID: 31.35 -> 33.54 (0.5); 32.97 (2)Hurt

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