NativeTok: Native Visual Tokenization for Improved Image Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911410994282496 |
|---|---|
| author | Wu, Bin Huang, Mengqi Jia, Weinan Mao, Zhendong |
| author_facet | Wu, Bin Huang, Mengqi Jia, Weinan Mao, Zhendong |
| contents | VQ-based image generation typically follows a two-stage pipeline: a tokenizer encodes images into discrete tokens, and a generative model learns their dependencies for reconstruction. However, improved tokenization in the first stage does not necessarily enhance the second-stage generation, as existing methods fail to constrain token dependencies. This mismatch forces the generative model to learn from unordered distributions, leading to bias and weak coherence. To address this, we propose native visual tokenization, which enforces causal dependencies during tokenization. Building on this idea, we introduce NativeTok, a framework that achieves efficient reconstruction while embedding relational constraints within token sequences. NativeTok consists of: (1) a Meta Image Transformer (MIT) for latent image modeling, and (2) a Mixture of Causal Expert Transformer (MoCET), where each lightweight expert block generates a single token conditioned on prior tokens and latent features. We further design a Hierarchical Native Training strategy that updates only new expert blocks, ensuring training efficiency. Extensive experiments demonstrate the effectiveness of NativeTok. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_22837 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | NativeTok: Native Visual Tokenization for Improved Image Generation Wu, Bin Huang, Mengqi Jia, Weinan Mao, Zhendong Computer Vision and Pattern Recognition VQ-based image generation typically follows a two-stage pipeline: a tokenizer encodes images into discrete tokens, and a generative model learns their dependencies for reconstruction. However, improved tokenization in the first stage does not necessarily enhance the second-stage generation, as existing methods fail to constrain token dependencies. This mismatch forces the generative model to learn from unordered distributions, leading to bias and weak coherence. To address this, we propose native visual tokenization, which enforces causal dependencies during tokenization. Building on this idea, we introduce NativeTok, a framework that achieves efficient reconstruction while embedding relational constraints within token sequences. NativeTok consists of: (1) a Meta Image Transformer (MIT) for latent image modeling, and (2) a Mixture of Causal Expert Transformer (MoCET), where each lightweight expert block generates a single token conditioned on prior tokens and latent features. We further design a Hierarchical Native Training strategy that updates only new expert blocks, ensuring training efficiency. Extensive experiments demonstrate the effectiveness of NativeTok. |
| title | NativeTok: Native Visual Tokenization for Improved Image Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.22837 |