AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Decoupling Complexity from Scale in Latent Diffusion Model

Tianxiong Zhong, Xingye Tian, Xuebo Wang, Boyuan Jiang, Xin Tao, Pengfei Wan

ECCV 2026 · Image generation › Autoregressive and latent generation · approach: Latent diffusion with learned tokenizers · 0 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
use entropy-guided content-adaptive token allocation across patches instead of uniform (ablation)DCS-Tok · ImageNet val reconstruction (avg 2 levels per patch)PSNR: 34.25 -> 34.99Improved
add L2 regularization with margin to latent denoising (ablation)DCS-Tok · Tokenizer ablation, generation fragmentationfragmentation / PSNR at 8 levels: 29.64 -> 32.61 PSNR at 8 levels; fragmentation removedImproved
add latent denoising task alone (no L2 regularization) (ablation)DCS-Tok · Tokenizer ablation, generationgeneration fragmentationNo clear effect
add L2 regularization without margin (ablation)DCS-Tok · Tokenizer ablation, reconstructionPSNR at 5-8 levels: 33.15 -> 29.64 at 8 levelsHurt
randomly drop tail token levels per patch during training (structured latents) (ablation)DCS-Tok · qualitative reconstructionhierarchy of latentsImproved
use linear level sampling probability instead of uniform (ablation)DCS-Tok · reconstructioncontribution of later-level tokensImproved
add more token levels in reconstruction (ablation)DCS-Tok · Tokenizer ablation (no denoise, no L2)PSNR: 24.22 -> 33.22 (1 -> 8 levels)Improved
DCS-LDM with first-level tokens vs state-of-the-art LDMs on ImageNetDCS-LDM · ImageNet 256x256 label-to-image (w/o CFG)gFID: 2.17 -> 7.68 (LightningDiT vs DCS-LDM)Hurt
DCS-LDM vs VFRTok on UCF101 video generationDCS-LDM · UCF101 24fps 256x256 label-to-video (w/ CFG)gFVD: 71.34 -> 86.87Hurt

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