Autoregressive Image Generation with Masked Bit Modeling

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yu, Qihang, Liu, Qihao, He, Ju, Zhang, Xinyang, Liu, Yang, Chen, Liang-Chieh, Chen, Xi
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917260987203584
author Yu, Qihang
Liu, Qihao
He, Ju
Zhang, Xinyang
Liu, Yang
Chen, Liang-Chieh
Chen, Xi
author_facet Yu, Qihang
Liu, Qihao
He, Ju
Zhang, Xinyang
Liu, Yang
Chen, Liang-Chieh
Chen, Xi
contents This paper challenges the dominance of continuous pipelines in visual generation. We systematically investigate the performance gap between discrete and continuous methods. Contrary to the belief that discrete tokenizers are intrinsically inferior, we demonstrate that the disparity arises primarily from the total number of bits allocated in the latent space (i.e., the compression ratio). We show that scaling up the codebook size effectively bridges this gap, allowing discrete tokenizers to match or surpass their continuous counterparts. However, existing discrete generation methods struggle to capitalize on this insight, suffering from performance degradation or prohibitive training costs with scaled codebook. To address this, we propose masked Bit AutoRegressive modeling (BAR), a scalable framework that supports arbitrary codebook sizes. By equipping an autoregressive transformer with a masked bit modeling head, BAR predicts discrete tokens through progressively generating their constituent bits. BAR achieves a new state-of-the-art gFID of 0.99 on ImageNet-256, outperforming leading methods across both continuous and discrete paradigms, while significantly reducing sampling costs and converging faster than prior continuous approaches. Project page is available at https://bar-gen.github.io/
format Preprint
id arxiv_https___arxiv_org_abs_2602_09024
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Autoregressive Image Generation with Masked Bit Modeling
Yu, Qihang
Liu, Qihao
He, Ju
Zhang, Xinyang
Liu, Yang
Chen, Liang-Chieh
Chen, Xi
Computer Vision and Pattern Recognition
This paper challenges the dominance of continuous pipelines in visual generation. We systematically investigate the performance gap between discrete and continuous methods. Contrary to the belief that discrete tokenizers are intrinsically inferior, we demonstrate that the disparity arises primarily from the total number of bits allocated in the latent space (i.e., the compression ratio). We show that scaling up the codebook size effectively bridges this gap, allowing discrete tokenizers to match or surpass their continuous counterparts. However, existing discrete generation methods struggle to capitalize on this insight, suffering from performance degradation or prohibitive training costs with scaled codebook. To address this, we propose masked Bit AutoRegressive modeling (BAR), a scalable framework that supports arbitrary codebook sizes. By equipping an autoregressive transformer with a masked bit modeling head, BAR predicts discrete tokens through progressively generating their constituent bits. BAR achieves a new state-of-the-art gFID of 0.99 on ImageNet-256, outperforming leading methods across both continuous and discrete paradigms, while significantly reducing sampling costs and converging faster than prior continuous approaches. Project page is available at https://bar-gen.github.io/
title Autoregressive Image Generation with Masked Bit Modeling
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2602.09024