通过空间局部性并行生成图像，显著提升自回归模型效率。
ZipAR: Parallel Auto-regressive Image Generation through Spatial Locality
- 利用图像局部结构特性，同时预测多个空间相邻区域的像素
- 在Emu3-Gen上减少91%前向传播次数，无需重新训练
- 即插即用，适合需要高效图像生成的应用场景
本文提出ZipAR，一种无需训练、可即插即用的并行解码框架，用于加速自回归视觉生成。其核心思想是图像具有局部结构，空间上相距较远的区域相互依赖性极低。在部分解码的基础上，除了传统的行方向逐个预测下一个令牌外，还可并行解码列方向上空间相邻区域的令牌，实现“下一组预测”范式。通过单次前向传播同时生成多个令牌，显著减少了生成完整图像所需的前向传播次数，极大提升了生成效率。实验表明，ZipAR在Emu3-Gen模型上可将前向传播次数减少高达91%，且无需任何额外训练。代码已开源：https://github.com/ThisisBillhe/ZipAR。
原文摘要 · Abstract (English)
In this paper, we propose ZipAR, a training-free, plug-and-play parallel decoding framework for accelerating auto-regressive (AR) visual generation. The motivation stems from the observation that images exhibit local structures, and spatially distant regions tend to have minimal interdependence. Given a partially decoded set of visual tokens, in addition to the original next-token prediction scheme in the row dimension, the tokens corresponding to spatially adjacent regions in the column dimension can be decoded in parallel, enabling the ``next-set prediction'' paradigm. By decoding multiple tokens simultaneously in a single forward pass, the number of forward passes required to generate an image is significantly reduced, resulting in a substantial improvement in generation efficiency. Experiments demonstrate that ZipAR can reduce the number of model forward passes by up to 91% on the Emu3-Gen model without requiring any additional retraining. Code is available here: https://github.com/ThisisBillhe/ZipAR.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。