跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 2026-10-09AI 評分36

MixTok: Adaptive Visual Tokenization with Global 1D Tokens and Sparse 2D Refinement

MixTok: Adaptive Visual Tokenization with Global 1D Tokens and Sparse 2D Refinement

AI 導讀

MixTok 結合精簡的 1D 全局表示與稀疏 2D token,按圖像內容自適應分配局部細化 token。於 ImageNet 256 x 256 上,重建 rFID 為 1.195,平均每張圖像約使用 128 個內容 token;生成 gFID 為 1.95,平均約使用 131 個內容 token。

正文

Published on Oct 9

Authors:

,

,

,

,

Abstract

Efficient visual tokenization requires balancing representation compactness with the preservation of fine-grained image content. Conventional grid-based tokenizers allocate tokens uniformly across spatial locations despite non-uniform visual complexity, whereas compact 1D tokenizers aggregate global information but can struggle to preserve local details. We introduce MixTok, a hybrid visual tokenizer that combines a compact 1D global representation with sparse 2D tokens for local refinement. A content-adaptive router ranks candidate locations by refinement priority and determines the number of 2D tokens for each image under a target average token budget. At the highest-scoring locations, 2D token features replace the spatial features recovered from the 1D representation, and a shared decoder reconstructs the image from the resulting mixed features. On ImageNet 256 x 256, MixTok achieves highly competitive reconstruction and generation performance, with an rFID of 1.195 and a gFID of 1.95 at respective average budgets of approximately 128 and 131 content tokens per image.

View arXiv page View PDF Project page GitHub Add to collection

Get this paper in your agent:

hf papers read 2610.13181

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2610.13181 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2610.13181 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.13181 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co