ACEdit：让多轮图像编辑在累积分布偏移下保持稳定的训练方案
ACEdit: Stable Multi-Turn Image Editing via Asymmetric Conditioning
多轮修图会越修越糊，这篇论文提出训练时让学生模型在自己输出上练习、教师模型锚定原图做蒸馏，长序列编辑明显更稳。
ACEdit 是一篇 arXiv 论文，针对图像编辑模型在多轮编辑中质量逐轮下降的问题。作者将 Self-Forcing 原则引入图像编辑，让模型在训练时就作用于自己上一轮的输出。由于自生成轨迹偏离 ground-truth 轨迹，直接监督不可靠，论文采用蒸馏目标，让教师编辑器锚定原始干净图像并接收复合编辑指令，学生则在自身输出上顺序编辑，构成非对称条件方案。实验显示长序列多轮编辑的退化显著减少，同时保持编辑质量和指令遵循。
ACEdit: Stable Multi-Turn Image Editing via Asymmetric Conditioning
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 through a distillation objective that leverages a teacher image editor to provide a reliable training signal when the model operates on its own previous outputs. We introduce an asymmetric conditioning scheme: while the student performs edits sequentially on its own outputs, the teacher remains anchored to the original clean image and receives a compound editing instruction. Experiments show substantially reduced degradation over long multi-turn sequences while maintaining editing quality and instruction adherence.
