AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Steering One-Step Diffusion Model with Fidelity-Rich Decoder for Fast Image Compression

Zheng Chen, Mingde Zhou, Jinpei Guo, Jiale Yuan, Yifei Ji, Yulun Zhang

AAAI 2026 · 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
add fidelity guidance module (aux. fidelity guidance) (ablation)SODEC · DIV2K-ValMS-SSIM: 0.8212 -> 0.8481Improved
add fidelity guidance module (ablation)SODEC · DIV2K-ValLPIPS: 0.3625 -> 0.3351Improved
use text prompt guidance instead of none (ablation)SODEC · DIV2K-ValMS-SSIM: 0.8212 -> 0.8185Hurt
use hyperprior guidance instead of none (ablation)SODEC · DIV2K-ValMS-SSIM: 0.8212 -> 0.8258Improved
add MSE alignment loss (ablation)SODEC · DIV2K-ValLPIPS: 0.4210 -> 0.3827Improved
use MSE+LPIPS composite alignment loss instead of MSE only (ablation)SODEC · DIV2K-ValMS-SSIM: 0.7948 -> 0.7481Hurt
merge alignment into main loss instead of separate term (ablation)SODEC · DIV2K-ValLPIPS: 0.3827 -> 0.4023Hurt
use rate annealing instead of frozen VAE (ablation)SODEC · DIV2K-ValMS-SSIM: 0.8512 -> 0.8951Improved
use rate annealing instead of joint training at matched bpp (ablation)SODEC · DIV2K-ValLPIPS: 0.3750 -> 0.3113Improved
use high-to-low rate annealing instead of low-to-high bpp curriculum (ablation)SODEC · DIV2K-ValMS-SSIM: 0.8643 -> 0.8951Improved
one-step diffusion decoding instead of multi-step PerCoSODEC · DIV2K-Valdecoding time (ms): 4,702.2 -> 227.9Improved
one-step diffusion vs VAE-only HiFiC decodingSODEC · DIV2K-Valdecoding time (ms): 3.9 -> 227.9Hurt

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