AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

VibeToken: Scaling 1D Image Tokenizers and Autoregressive Models for Dynamic Resolution Generations

Maitreya Patel, Jingtao Li, Weiming Zhuang, Yezhou Yang, Lingjuan Lv

CVPR 2026 · Image generation › Autoregressive and latent generation · approach: Next-scale and token autoregressive images · 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
add dynamic grid position embedding in place of learnable RoPE (ablation)TiTok-S-128 VAE tokenizer · VAE small-scale study, 1024x1024rFID: 1.544 -> 1.707 (256), 93.49 -> 131.20 (1024)Mixed
add adaptive patch embedding on top of dynamic grid embedding (ablation)TiTok-S-128 VAE tokenizer · VAE small-scale study, 1024x1024rFID: 131.20 -> 5.38Improved
add plain RoPE to the TiTok baseline (ablation)TiTok-S-128 VAE tokenizer · VAE small-scale study, 256x256rFID: 1.862 -> 2.934Hurt
add learnable RoPE to the TiTok baseline (ablation)TiTok-S-128 VAE tokenizer · VAE small-scale study, 256x256rFID: 1.862 -> 1.544Improved
use 64 tokens instead of 256 for the AR generator (ablation)VibeToken-Gen GPT-B · ImageNet 256 gFID with cfggFID: 9.81 -> 8.42Improved
use 128 tokens instead of 256 for the AR generator (ablation)VibeToken-Gen GPT-B · ImageNet 256 gFID with cfggFID: 9.81 -> 9.02Improved
train as a mixed-resolution generalist with 64 tokens (ablation)VibeToken-Gen GPT-B · ImageNet 256 gFIDgFID: w/o cfg 33.15 -> 31.14, w/ cfg 8.42 -> 9.37Mixed
train as a mixed-resolution generalist with 128 tokens (ablation)VibeToken-Gen GPT-B · ImageNet 256 gFIDgFID: w/o cfg 37.68 -> 39.50, w/ cfg 9.02 -> 10.58Hurt
replace LlamaGen (576 tokens, fixed resolution) with VibeToken-Gen (64 tokens, generalist)GPT-B AR generator · ImageNet 256 gFIDgFID: w/o cfg 33.44 -> 31.14, w/ cfg 7.15 -> 9.37Mixed
replace diffusion NiT with VibeToken-Gen-XXL for 1024x1024 generationclass-conditional generator · ImageNet 1024x1024gFID: 5.87 -> 3.54Improved
scale VibeToken-Gen from B to XXLVibeToken-Gen · Low-resolution average gFIDgFID: 9.14 -> 5.63Improved
use VibeToken-LL (large encoder) instead of VibeToken-SLVibeToken tokenizer · Arbitrary-resolution reconstructionrFID: 4.21 -> 3.60Improved

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