AnchorEdit: 通过因果记忆在多轮图像编辑中保持时间一致性
AnchorEdit: Maintaining Temporal Consistency in Multi-turn Image Editing via Causal Memory
AI总结：
提出首个自回归扩散框架AnchorEdit，通过因果记忆机制和自展开策略解决多轮编辑中的身份漂移和误差累积问题，在10轮以上交互中保持高保真度。
摘要：
多轮图像编辑对于迭代设计至关重要，但当前模型在连续步骤中常面临身份漂移和误差累积。现有研究利用视频先验保持一致性，但其依赖的双向注意力与交互式编辑的因果、顺序性质根本不符。本文提出AnchorEdit，首个专为高分辨率、长期多轮编辑设计的自回归（AR）扩散框架。AnchorEdit通过三阶段训练课程弥合视频先验与因果推理之间的差距：保持身份的单轮预训练、使用新颖的自展开策略进行因果AR强制微调以缓解暴露偏差，以及用于高效4步生成的一致性蒸馏。在推理过程中，我们引入记忆机制来锚定初始主体身份，并确保在扩展编辑轨迹上的稳定外推。为评估性能，我们提供了一个新的高分辨率多轮编辑基准，旨在压力测试长期稳定性。大量实验表明，AnchorEdit达到了最先进的结果，即使在10轮以上的交互中也能保持卓越的主体保真度和指令遵循能力。
英文摘要：
Multi-turn image editing is essential for iterative design, yet current models often struggle with identity drift and error accumulation over successive steps. While existing research leverages video priors for consistency, their reliance on bidirectional attention is fundamentally misaligned with the causal, sequential nature of interactive editing. In this paper, we propose AnchorEdit, the first autoregressive (AR) diffusion-based framework designed specifically for high-resolution, long-term multi-turn editing. AnchorEdit bridges the gap between video priors and causal inference through a three-stage training curriculum: identity-preserving sing-turn pretraining, causal AR forcing fine-tuning with a novel self-rollout strategy to mitigate exposure bias, and consistency distillation for efficient 4-step generation. During inference, we introduce a memory mechanism to anchor the initial subject identity and ensure stable extrapolation across extended editing trajectories. To evaluate performance, we provide a new high-resolution multi-turn editing benchmark designed to stress-test long-horizon stability. Extensive experiments demonstrate that AnchorEdit achieves state-of-the-art results, maintaining exceptional subject fidelity and instruction following even over 10+ interaction rounds.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
