ResTok: Learning Hierarchical Residuals in 1D Visual Tokenizers for Autoregressive Image Generation

TL;DR

ResTok introduces a hierarchical residual approach to 1D visual tokenization, significantly improving autoregressive image generation efficiency and quality by capturing multi-level features and reducing sampling steps.

Contribution

The paper proposes ResTok, a novel hierarchical residual visual tokenizer that enhances representation and accelerates autoregressive image generation.

Findings

Achieves a gFID of 2.34 on ImageNet-256 with 9 sampling steps.

Significantly improves autoregressive image generation quality.

Reduces sampling steps by predicting entire latent levels at once.

Abstract

Existing 1D visual tokenizers for autoregressive (AR) generation largely follow the design principles of language modeling, as they are built directly upon transformers whose priors originate in language, yielding single-hierarchy latent tokens and treating visual data as flat sequential token streams. However, this language-like formulation overlooks key properties of vision, particularly the hierarchical and residual network designs that have long been essential for convergence and efficiency in visual models. To bring "vision" back to vision, we propose the Residual Tokenizer (ResTok), a 1D visual tokenizer that builds hierarchical residuals for both image tokens and latent tokens. The hierarchical representations obtained through progressively merging enable cross-level feature fusion at each layer, substantially enhancing representational capacity. Meanwhile, the semantic residuals between hierarchies prevent information overlap, yielding more concentrated latent distributions that are easier for AR modeling. Cross-level bindings consequently emerge without any explicit constraints. To accelerate the generation process, we further introduce a hierarchical AR generator that substantially reduces sampling steps by predicting an entire level of latent tokens at once rather than generating them strictly token-by-token. Extensive experiments demonstrate that restoring hierarchical residual priors in visual tokenization significantly improves AR image generation, achieving a gFID of 2.34 on ImageNet-256 with only 9 sampling steps. Code is available at https://github.com/Kwai-Kolors/ResTok.