ACEdit: Stable Multi-Turn Image Editing via Asymmetric Conditioning
Summary
ACEdit 提出一种结合自强制（Self-Forcing）训练、教师蒸馏与非对称条件化的方案，使多轮图像编辑在累积分布偏移下保持稳定，显著降低长序列编辑中的质量退化。
View Cached Full Text
Cached at: 10/21/26, 07:06 AM
Paper page - ACEdit: Stable Multi-Turn Image Editing via Asymmetric Conditioning
Source: https://huggingface.co/papers/2610.22328
Abstract
Self-forcing training with teacher distillation and asymmetric conditioning stabilizes multi-turn image editing against accumulating distribution shifts.
Image editing models produce impressive results for individual edits, but their quality often deteriorates over multiple editing turns. Each editing turn introduces smalldistribution 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 makesmulti-turn image editingstable under this accumulatingdistribution shift. A key is to force the editor to operate on its own outputs from previous editing turns during training, bringing the principle ofSelf-Forcingto 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 adistillation objectivethat leverages a teacher image editor to provide a reliable training signal when the model operates on its own previous outputs. We introduce anasymmetric conditioningscheme: 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.
View arXiv pageView PDFProject pageAdd to collection
Get this paper in your agent:
hf papers read 2610\.22328
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2610.22328 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2610.22328 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2610.22328 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
MT Forcing: Towards Stable and Consistent Multi-Turn Image Editing with Autoregressive Diffusion
论文提出 MT Forcing，通过单轮分布匹配蒸馏与教师强制引导，提升基于自回归扩散的统一多模态模型在多轮图像编辑中的视觉质量稳定性与跨轮内容一致性。
MTEdit: Packed Trajectories Beat Rollouts for Multi-Turn Image Editing
MTEdit introduces a 500K multi-turn image editing dataset built with packed trajectory canvas generation to reduce visual drift, plus MTEdit-Bench, and shows that fine-tuning Qwen-Image-Edit-Causal yields the best open-source multi-turn editing performance.
On-Policy Self-Distillation for Multi-Turn Image Editing
This paper proposes MT-OPSD, a non-policy self-distillation framework to address degradation in multi-turn image editing and introduces LME-Bench for evaluating long-horizon robustness.
HP-Edit: A Human-Preference Post-Training Framework for Image Editing
HP-Edit introduces a post-training framework that aligns diffusion-based image editing models with human preferences via RLHF, using a new 50K real-world dataset and an automatic VLM-based evaluator.
Uni-Edit: Intelligent Editing Is A General Task For Unified Model Tuning
Uni-Edit proposes using intelligent image editing as a single general task to simultaneously improve unified multimodal models' understanding, generation, and editing capabilities, with an automated data synthesis pipeline creating complex editing instructions.