paper-with-me

홈 › Papers

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

2025-07-07 · Yecheng Wu, Junyu Chen, Zhuoyang Zhang, Enze Xie, Jincheng Yu, Junsong Chen, Jinyi Hu, Yao Lu, Song Han, Han Cai

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.

📄 PDF Abstract BibTeX arXiv:2507.04947

Code (0)

등록된 구현이 없습니다.

Tasks

Computational EfficiencyImage GenerationText to Image GenerationText-to-Image Generation

Similar Papers 제목 키워드 기반

PerCoV2: Improved Ultra-Low Bit-Rate Perceptual Image Compression with Implicit Hierarchical Masked Image Modeling

2025-03-12 · Nikolai Körber, Eduard Kromer, Andreas Siebert, Sascha Hauke 외

We introduce PerCoV2, a novel and open ultra-low bit-rate perceptual image compression system designed for bandwidth- and storage-constrained applications. Building upon prior work by Careil et al., PerCoV2 extends the o…

Image Compression

Taming Teacher Forcing for Masked Autoregressive Video Generation

2025-01-21 · CVPR 2025 1 · Deyu Zhou, Quan Sun, Yuang Peng, Kun Yan 외

We introduce MAGI, a hybrid video generation framework that combines masked modeling for intra-frame generation with causal modeling for next-frame generation. Our key innovation, Complete Teacher Forcing (CTF), conditio…

Video GenerationVideo Prediction

Autoregressive Image Generation with Masked Bit Modeling

2026-02-09 · Qihang Yu, Qihao Liu, Ju He, Xinyang Zhang 외 arxiv

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…

Image Generation

MaskMamba: A Hybrid Mamba-Transformer Model for Masked Image Generation

2024-09-30 · Wenchao Chen, LiQiang Niu, Ziyao Lu, Fandong Meng 외

Image generation models have encountered challenges related to scalability and quadratic complexity, primarily due to the reliance on Transformer-based backbones. In this study, we introduce MaskMamba, a novel hybrid mod…

Image GenerationMambaText to Image GenerationText-to-Image Generation

Multi-scale and Context-adaptive Entropy Model for Image Compression

2019-10-17

We propose an end-to-end trainable image compression framework with a multi-scale and context-adaptive entropy model, especially for low bitrate compression. Due to the success of autoregressive priors in probabilistic g…

Image CompressionMS-SSIMSSIM