DC-AR: Efficient Masked Autoregressive Image Generation with Deep Compression Hybrid Tokenizer

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Wu, Yecheng, Chen, Junyu, Zhang, Zhuoyang, Xie, Enze, Yu, Jincheng, Chen, Junsong, Hu, Jinyi, Lu, Yao, Han, Song, Cai, Han
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911042442887168
author Wu, Yecheng
Chen, Junyu
Zhang, Zhuoyang
Xie, Enze
Yu, Jincheng
Chen, Junsong
Hu, Jinyi
Lu, Yao
Han, Song
Cai, Han
author_facet Wu, Yecheng
Chen, Junyu
Zhang, Zhuoyang
Xie, Enze
Yu, Jincheng
Chen, Junsong
Hu, Jinyi
Lu, Yao
Han, Song
Cai, Han
contents We introduce DC-AR, a novel masked autoregressive (AR) text-to-image generation framework that delivers superior image generation quality with exceptional computational efficiency. Due to the tokenizers' limitations, prior masked AR models have lagged behind diffusion models in terms of quality or efficiency. We overcome this limitation by introducing DC-HT - a deep compression hybrid tokenizer for AR models that achieves a 32x spatial compression ratio while maintaining high reconstruction fidelity and cross-resolution generalization ability. Building upon DC-HT, we extend MaskGIT and create a new hybrid masked autoregressive image generation framework that first produces the structural elements through discrete tokens and then applies refinements via residual tokens. DC-AR achieves state-of-the-art results with a gFID of 5.49 on MJHQ-30K and an overall score of 0.69 on GenEval, while offering 1.5-7.9x higher throughput and 2.0-3.5x lower latency compared to prior leading diffusion and autoregressive models.
format Preprint
id arxiv_https___arxiv_org_abs_2507_04947
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DC-AR: Efficient Masked Autoregressive Image Generation with Deep Compression Hybrid Tokenizer
Wu, Yecheng
Chen, Junyu
Zhang, Zhuoyang
Xie, Enze
Yu, Jincheng
Chen, Junsong
Hu, Jinyi
Lu, Yao
Han, Song
Cai, Han
Computer Vision and Pattern Recognition
Artificial Intelligence
We introduce DC-AR, a novel masked autoregressive (AR) text-to-image generation framework that delivers superior image generation quality with exceptional computational efficiency. Due to the tokenizers' limitations, prior masked AR models have lagged behind diffusion models in terms of quality or efficiency. We overcome this limitation by introducing DC-HT - a deep compression hybrid tokenizer for AR models that achieves a 32x spatial compression ratio while maintaining high reconstruction fidelity and cross-resolution generalization ability. Building upon DC-HT, we extend MaskGIT and create a new hybrid masked autoregressive image generation framework that first produces the structural elements through discrete tokens and then applies refinements via residual tokens. DC-AR achieves state-of-the-art results with a gFID of 5.49 on MJHQ-30K and an overall score of 0.69 on GenEval, while offering 1.5-7.9x higher throughput and 2.0-3.5x lower latency compared to prior leading diffusion and autoregressive models.
title DC-AR: Efficient Masked Autoregressive Image Generation with Deep Compression Hybrid Tokenizer
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2507.04947