| title | Channel-wise Vector Quantization |
|---|---|
| authors | Wei Song, Tianhang Wang, Yitong Chen, Tong Zhang, Zuxuan Wu, Ming Li, Jiaqi Wang, Kaicheng Yu |
| date | 2026-05-25 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2605.26089 |
| upvotes | 13 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| abstract | We present Channel-wise Vector Quantization (CVQ), a novel image tokenization paradigm that replaces patch-wise tokens with channel-wise tokens. Unlike conventional vector quantization, which assigns a discrete token to each patch feature vector, CVQ quantizes each channel of the feature map. This formulation represents an image as discrete levels of visual details, rather than as a grid of spatial patches. Based on CVQ, we introduce a new visual autoregressive framework with "next-channel prediction". Instead of rendering images patch by patch in raster order, our Channel-wise Autoregressive (CAR) model predicts image channels sequentially, producing progressively enriched visual details. Specifically, it first sketches global structure and then refines fine-grained attributes, akin to a human artist's workflow. Empirically, we show that: (1) CVQ achieves 100% codebook utilization with a 16K+ codebook size without any bells and whistles, and substantially improves reconstruction quality over conventional VQ; and (2) CAR attains a DPG score of 86.7 and a GenEval score of 0.79, demonstrating strong effectiveness for text-to-image generation. |
cite
@misc{song2026channel,
title={Channel-wise Vector Quantization},
author={Wei Song and Tianhang Wang and Yitong Chen and Tong Zhang and Zuxuan Wu and Ming Li and Jiaqi Wang and Kaicheng Yu},
year={2026},
eprint={2605.26089},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2605.26089}
}