只重绘需要修改的区域，提升编辑效率与质量。
NEP: Autoregressive Image Editing via Next Editing Token Prediction
- 基于自回归生成，仅预测需编辑区域的像素。
- 在多个基准上达到新最好性能，支持零样本编辑。
- 无需微调即可通过迭代优化持续改进结果。
文本引导的图像编辑通常只需修改源图像的小范围区域，但现有方法往往重建整个目标图像，导致计算开销大且易错误地复现非编辑区域，影响编辑质量。为此，我们提出基于自回归图像生成的下一编辑令牌预测（NEP）框架，仅对需编辑区域进行重生成，避免对非编辑区域的意外修改。为实现任意区域编辑，我们预训练一个任意顺序自回归文本到图像（T2I）模型。该模型训练完成后可直接用于零样本图像编辑，并轻松适配至NEP框架，在广泛使用的图像编辑基准上达到新的最优表现。此外，模型通过测试时缩放（TTS）机制，可零样本迭代优化生成结果，持续提升质量。
原文摘要 · Abstract (English)
Text-guided image editing involves modifying a source image based on a language instruction and, typically, requires changes to only small local regions. However, existing approaches generate the entire target image rather than selectively regenerate only the intended editing areas. This results in (1) unnecessary computational costs and (2) a bias toward reconstructing non-editing regions, which compromises the quality of the intended edits. To resolve these limitations, we propose to formulate image editing as Next Editing-token Prediction (NEP) based on autoregressive image generation, where only regions that need to be edited are regenerated, thus avoiding unintended modification to the non-editing areas. To enable any-region editing, we propose to pre-train an any-order autoregressive text-to-image (T2I) model. Once trained, it is capable of zero-shot image editing and can be easily adapted to NEP for image editing, which achieves a new state-of-the-art on widely used image editing benchmarks. Moreover, our model naturally supports test-time scaling (TTS) through iteratively refining its generation in a zero-shot manner. The project page is: https://nep-bigai.github.io/
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。