AI Paper Atlas

Research map / Image generation / Image restoration

CoD-Lite: Real-Time Diffusion-Based Generative Image Compression

Zhaoyang Jia, Naifu Xue, Zihan Zheng, Jiahao Li, Bin Li, Xiaoyi Zhang, Zongyu Guo, Yuan Zhang, Houqiang Li, Yan Lu

ICML 2026 · Image generation › Image restoration · approach: Perceptual diffusion image compression · 0 citations (OpenAlex)

This paper tried a common change: Train longer

train longer · Kodak 0.0312 bpp · FID: 32.8 -> 31.5 Improved

What 387 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

Read the paper

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
unconditional generation-oriented diffusion pre-training for large codec (ablation)700M diffusion codec · ImageNet 256, 0.0156 bppFID: 23.7 -> 15.4Improved
unconditional generation-oriented diffusion pre-training for small codec (ablation)34M diffusion codec · ImageNet 256, 0.0156 bppFID: 2.0 FID gain vs 8.3 for large modelNo clear effect
class-conditioned pre-training instead of unconditional for small model (ablation)34M diffusion codec · ImageNet 256, 0.0156 bppFIDImproved
class-conditioned pre-training instead of unconditional for large model (ablation)700M diffusion codec · ImageNet 256, 0.0156 bppFIDHurt
use compression-oriented CoD pre-training instead of generation-oriented (ablation)34M diffusion codec · ImageNet 256, 0.0156 bppFID: -3.5 FID (34M); -3.4 FID (700M)Improved
add CoD pre-training to from-scratch baseline (ablation)one-step codec · Kodak 0.0312 bppFID: 41.1 -> 29.6Improved
replace global attention with local window attention in CoD pre-training (ablation)CoD · Kodak 0.0039 bpp pre-trainFID: 59.7 -> 60.2No clear effect
replace attention by depth-wise conv in CoD pre-training (ablation)CoD · Kodak 0.0039 bpp pre-trainFID: 59.7 -> 62.5Hurt
replace attention by depth-wise conv in one-step codec (ablation)one-step DiT codec · Kodak 0.0312 bppFID and decode speed: FID 37.5 -> 41.4; 331.2 ms -> 23.6 msMixed
add DMD distillation loss from large CoD teacher (ablation)conv one-step codec · Kodak 0.0312 bppFID: 41.4 -> 36.4Improved
add DMD distillation loss to DiT global attention codec (ablation)DiT one-step codec · Kodak 0.0312 bppFID: 37.5 -> 35.4Improved
replace PatchGAN with projected GAN (DINOv2 features) (ablation)CoD-Lite · Kodak 0.0312 bppFID: 36.4 -> 32.8Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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