AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

LacTokGen: Latent Consistency Tokenizer for 1024-pixel Image Generation by 256 Tokens

Qingsong Xie, Luyuan Zhang, Zhao Zhang, Siyuan Li, Zhe Huang, Zhenyu Yang, Haonan Lu

CVPR 2026 · Image generation › Autoregressive and latent generation · approach: Generative latent space tooling

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 pixel reconstruction loss with consistency acceleration (LacTok-H) over VQ-LADD with diffusion loss only (ablation)LacTok · MJHQ-5K 1024px reconstructionrFID: 12.72 -> 11.34Improved
swap LlamaGen decoder for latent diffusion decoder trained with diffusion loss only (ablation)LlamaGen tokenizer · MJHQ-5K 1024px reconstructionPSNR: 19.24 -> 16.53Hurt
add HyperSD or TLCM acceleration without pixel loss to VQ-LADD (ablation)VQ-LADD · MJHQ-5K 1024px reconstructionrFID: 12.72 -> 14.71 (HyperSD), 14.40 (TLCM)Hurt
use TLCM (two-step) instead of HyperSD (one-step) in the decoder (ablation)LacTok with constructed data · FLUX-5K 1024px reconstructionrFID: 12.18 -> 11.04Improved
train tokenizer on constructed FLUX-synthesized data instead of ImageNet (ablation)LacTok-H · MSCOCO-2017 5K reconstructionrFID: 10.80 -> 16.93Hurt
train tokenizer on constructed data (ablation)LacTok-H · FLUX-5K reconstructionrFID: 12.45 -> 12.18Improved
increase token count from 192 to 324 (ablation)LacTok-H · MJHQ-5K 1024px reconstructionrFID: 12.35 -> 11.01Improved
train autoregressive model on constructed high-quality data (ablation)LacTokGen · GenEvaloverall score: 0.71 -> 0.72Improved
use TLCM instead of HyperSD in LacTokGen (ablation)LacTokGen · GenEvaloverall score: 0.72 -> 0.73Improved
increase CFG scale above 2 (ablation)LacTokGen-L* · MSCOCO-2017 5K generationHPSv2: 0.303 (1.5) -> 0.304 (2, 3, 7)No clear effect

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