AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

AToken: A Unified Tokenizer for Vision

Jiasen Lu, Liangchen Song, Mingze Xu, Byeongjoo Ahn, Yanjun Wang, Chen Chen, Afshin Dehghan, Yinfei Yang

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 video training (Stage 2) to image-only tokenizer (ablation)AToken-So/C (800M) · ImageNet reconstructionrFID: 0.26 -> 0.25Improved
add 3D training (Stage 3) after image and video (ablation)AToken-So/C (800M) · ImageNet reconstructionrFID: 0.25 -> 0.21Improved
add 3D training (Stage 3) (ablation)AToken-So/C (800M) · TokenBench video reconstructionrFVD: 3.63 -> 3.01Improved
use smaller Base (192M) model instead of So400m (800M) under multimodal expansion (ablation)AToken · ImageNet rFID across stagesrFID: 0.323 -> 0.483Hurt
use Gram matrix loss instead of GAN loss for reconstruction (ablation)AToken transformer tokenizer · ImageNet reconstructionrFIDImproved
add video stage before 3D stage for generation tokenizer (ablation)AToken-So/C with Lightning-DiT · ImageNet generation (continuous)gFID: 1.62 -> 1.88Hurt
add 3D stage (Stage 3) after video stage (ablation)AToken-So/C with Lightning-DiT · ImageNet generation (continuous)gFID: 1.88 -> 1.56Improved
quantize latents with FSQ (discrete variant)AToken-So · ImageNet reconstructionrFID: 0.21 -> 0.38Hurt
replace Oryx-ViT with frozen AToken-So/C encoder in SlowFast-LLaVA-1.5-7BSlowFast-LLaVA-1.5-7B · image understanding (7 benchmarks)accuracy: RW-QA 67.5 -> 68.8; OCRBench 76.4 -> 74.5Mixed
replace Oryx-ViT with frozen AToken-So/C encoder in SlowFast-LLaVA-1.5-7BSlowFast-LLaVA-1.5-7B · video understandingaccuracy: VideoMME 63.9 -> 64.5; LongVideoBench 62.5 -> 60.6Mixed

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