AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

EvoTok: A Unified Image Tokenizer via Residual Latent Evolution for Visual Understanding and Generation

Yan Li, Ning Liao, Xiangyu Zhao, Shaofeng Zhang, Xiaoxing Wang, Yifan Yang, Junchi Yan, Xue Yang

ECCV 2026 · Image generation › Autoregressive and latent generation · approach: Generative latent space tooling · 2 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 pixel-to-semantic evolution (Lpix=4, Lsem=16) instead of entangled equal depth (4,4) (ablation)EvoTok · SEEDBenchaccuracy: 62.7 -> 67.1Improved
use pixel-to-semantic evolution instead of entangled equal depth (ablation)EvoTok · MMEscore: 1668.9 -> 1793.5Improved
use pixel-to-semantic evolution instead of entangled equal depth (ablation)EvoTok · GenEvaloverall: 0.64 -> 0.67Improved
use pixel-to-semantic evolution instead of entangled equal depth (ablation)EvoTok · ImageNet-1K valrFID: 0.66 -> 0.55Improved
use pixel-to-semantic evolution instead of semantic-to-pixel (16,4) (ablation)EvoTok · MMEscore: 1731.6 -> 1793.5Improved
use pixel-to-semantic evolution instead of semantic-to-pixel (ablation)EvoTok · GenEvaloverall: 0.60 -> 0.67Improved
use pixel-to-semantic evolution instead of semantic-to-pixel (ablation)EvoTok · ImageNet-1K valrFID: 0.44 -> 0.55Hurt
use pixel-to-semantic evolution instead of semantic-to-pixel (ablation)EvoTok · SEEDBenchaccuracy: 64.6 -> 67.1Improved
use pixel-to-semantic evolution instead of layer-wise decoupled features (ablation)EvoTok · ImageNet-1K valrFID: 0.59 -> 0.55Improved
use shared-latent evolution instead of separate encoder branches (ablation)EvoTok · ImageNet-1K valrFID: 0.68 -> 0.55Improved
use EvoTok instead of UniTok trained on billion-scale dataEvoTok · ImageNet-1K valrFID: 0.33 -> 0.43Hurt
use EvoTok instead of TokenFlow-B for understandingEvoTok · MMEscore: 1660.4 -> 1895.1Improved

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