Refinement Is Inherently Editable: Training-Free Prompt-to-Prompt Image Editing with Generative Refinement Network
- 类型：arxiv
- 标识：2609.20633
- 链接：https://arxiv.org/abs/2609.20633
- 主分类：engineering
- 形态：method
- TLDR：Text-guided image editing must introduce the requested changes while preserving unrelated source content. Diffusion-based editors rely on spatial controls whose inaccuracies can leave edits incomplete or alter unrelated regions. Causal autoregressive editors face a further constraint: their fixed decoding order limits revision of earlier decisions. We introduce RefineEdit, a training-free prompt-to-prompt image editing framework built on a Generative Refinement Network. Our key idea is to couple edit localization with content generation through the global refinement of binary image codes, allo
- 副分类：multimodal
- 待LLM分类：否
- 标题中文：精化本身即可编辑：基于生成式精化网络的无训练 Prompt-to-Prompt 图像编辑
- TLDR中文：文本引导的图像编辑需要在引入所请求修改的同时保留无关源内容。基于扩散的编辑器依赖空间控制，其不准确性可能导致编辑不完整或改变无关区域。因果自回归编辑器面临进一步限制：其固定解码顺序约束了对早期决策的修订。我们提出 RefineEdit，一个基于 Generative Refinement Network 的无训练 prompt-to-prompt 图像编辑框架。其核心思想是通过对二元图像码的全局精化，将编辑定位与内容生成耦合，从而允许
- 来源文件：
- /inbox/tom/_candidates/2026-09-21-agent-rag-longcontext-candidates.json