[论文] ACEdit: Stable Multi-Turn Image Editing via Asymmetric Conditioning

论文概要

研究领域: CV 作者: Sara Dorfman, Nir Goren, Gal Barak, Daniel Cohen-Or, Or Patashnik 发布时间: 2026-10-15 arXiv: 2610.22328

中文摘要

图像编辑模型对单次编辑产生令人印象深刻的结果，但其质量通常在多轮编辑中恶化，因为每轮编辑引入微小的分布偏移在连续编辑中累积。我们引入一种训练方案，使多轮图像编辑在累积分布偏移下保持稳定，迫使编辑器在训练期间操作自己前几轮的输出，将 Self-Forcing 原则引入图像编辑。主要挑战是自生成的编辑轨迹偏离真值轨迹，使直接监督不可靠。我们通过蒸馏目标解决：利用教师图像编辑器在模型操作自己先前输出时提供可靠的训练信号。我们引入非对称条件方案：学生在自身输出上顺序执行编辑，而教师始终锚定到原始干净图像并接收复合编辑指令。实验表明，该方法在长多轮序列中显著减少退化，同时保持编辑质量和指令遵循性。

原文摘要

Image editing models produce impressive results for individual edits, but their quality often deteriorates over multiple editing turns. Each editing turn introduces small distribution shifts that accumulate over successive edits, creating a gap between the input distributions encountered during the editor's original training and multi-turn inference. We introduce a training scheme that makes multi-turn image editing stable under this accumulating distribution shift. A key is to force the editor to operate on its own outputs from previous editing turns during training, bringing the principle of Self-Forcing to image editing. The main challenge is that self-generated editing trajectories deviate from ground-truth trajectories, making direct supervision unreliable. We address this challenge t...

#论文 #arXiv #CV #小凯