一次前向传播完成多区域图像编辑，避免属性混淆。
Editing Everything Everywhere All at Once
- 通过调节注意力机制实现多实例同时编辑
- 在8.5个编辑指令下仍保持视觉一致性
- 无需训练，适合需要快速批量编辑的场景
在单次前向传播中同时编辑图像多个元素，比多轮操作更高效且可能提升整体协调性。然而，当多个指令作用于不同区域时，语义干扰常导致属性泄露和编辑不分离，尤其在编辑数量增加时更为严重。本文提出MICE（多实例并发编辑），一种基于多模态扩散变换器的免训练策略，通过调整联合注意力中的加性偏置，调控实例特定编辑指令、潜在表示与上下文标记之间的交互。具体而言，MICE允许实例内部注意力，惩罚相邻区域标记间的交互，并抑制无关实例间的跨注意力。结果实现属性绑定并维持全局视觉一致性。我们在LoMOE-Bench上评估MICE，引入更难的MICE-Bench基准，平均每个图像包含8.5个并发编辑。实验表明，该方法在视觉质量保持和编辑指令忠实度方面优于强基线及近期竞争方法。
原文摘要 · Abstract (English)
Editing multiple elements of an image in a single forward pass is a practical alternative to multi-turn image manipulation, offering improved efficiency and potentially better harmonization. However, when several instructions target different regions, semantic interference often leads to attribute leakage and poor edit disentanglement, especially as the number of edits increases. In this work, we propose MICE (Multi-Instance Concurrent Editing), a training-free strategy for scalable multi-instance image editing with Multimodal Diffusion Transformers. MICE modifies the additive bias of joint attention to regulate interactions between instance-specific edit instructions, latent, and context tokens identified via user-provided segmentation masks. Specifically, MICE allows intra-instance attention, penalizes interactions between neighboring region tokens, and suppresses unrelated cross-instance attention. As a result, our method enforces attribute binding while preserving global visual consistency. We evaluate MICE on LoMOE-Bench and introduce MICE-Bench, a more challenging benchmark with an average of 8.5 concurrent edits per image. The experiments demonstrate that our approach outperforms strong baselines and recent competitors in terms of visual quality preservation and faithfulness to the editing instructions.
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。