内联批评引导图像编辑
Inline Critic Steers Image Editing
AI总结：
本文提出内联批评方法，通过在图像编辑过程中实时修正模型输出，提升生成质量，在多个基准测试中取得优异成绩。
摘要：
基于指令的图像编辑在不同案例和图像区域具有异质难度，推动了针对模型困难区域的细化方法。现有细化信号在完全生成图像或去噪步骤后才出现。本文探讨是否可以在持续的前向传递中使用此类信号。通过冻结图像编辑模型，发现尽管生成能力仅在最后几层出现，但误差模式已在早期层设定（rank correlation ρ = 0.83与最终层误差图）。基于此，本文引入内联批评，一种可学习的标记，可在冻结模型的中间层批评其预测并引导其隐藏状态以在前向传递中细化生成。提出三阶段配方以稳定训练，从学习批评到引导生成。结果在GEdit-Bench上达到SOTA（7.89），在RISEBench上比相同backbone提升9.4%，并在KRIS-Bench上取得最强开源结果（81.92，超越GPT-4o）。进一步分析显示，批评确实塑造了模型后续层的注意力和预测更新。
英文摘要：
Instruction-based image editing exhibits heterogeneous difficulty not only across cases but also across regions of an image, motivating refinement approaches that allocate correction to where the model struggles. Existing refinement signals arrive late, after a fully generated image or a completed denoising step. We ask whether such a signal can act within an ongoing forward pass. To investigate this, we probe a frozen image-editing model and find that although generation capability emerges only in the last few layers, the error pattern is already set in early layers (rank correlation \r{ho} = 0.83 with the final-layer error map). Based on this, we introduce Inline Critic, a learnable token that critiques a frozen model's predictions at its intermediate layers and steers its hidden states to refine generation during the forward pass. A three-stage recipe is proposed to stabilize the training from learning how to critique to steering generation. As a result, we achieve state of the art on GEdit-Bench (7.89), a +9.4 gain on RISEBench over the same backbone, and the strongest open-source result on KRIS-Bench (81.92, surpassing GPT-4o). We further provide analyses showing that the critic genuinely shapes the model's attention and prediction updates at subsequent layers.
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。
