把编辑结果拆成可复用的图层，AI 修图终于能改局部了
现在的 AI 修图是「一锤子买卖」：你让它把照片里的杯子换成红色，它给你一张改好的整图，想再单独调杯子的透明度、或者把杯子抠出来贴到另一张图上，做不到。这篇把编辑结果拆成一层层可复用的 RGBA 图层——背景一层、杯子一层、光影一层——你改完还能单独动其中任何一层，就像在 Photoshop 里操作。做法是把「编辑指令」和「图层生成」放进同一个模型流程里，再用一种蒸馏技巧保住原有编辑能力。它不是你明天就能用的工具，但这是 AI 修图从「出图」走向「可编辑资产」的关键一步。
📄 原文摘要(英文)
High-quality image editing requires precise modification of target regions while preserving the content of unrelated regions. Existing editing models produce flattened outputs that limit precise local manipulation and layer reuse, while decomposition methods typically optimize layer recovery separately from instruction-driven editing. To this end, we introduce LayerEditor, a unified framework that couples instruction-guided image editing with controllable generation of reusable RGBA layers. At its core, In-Context RGB-Alpha Token Modeling jointly represents the edited canvas and an ordered sequence of RGB-Alpha layers within a single token sequence, enabling semantic edits and explicit control over individual layer content. To retain the pretrained model's editing capability during adaptation to layered outputs, we propose Compositional Edit-Prior Distillation, which transfers its editing prior to both the edited canvas and the image composed from the predicted layers. To support training and evaluation, we construct LayerEditor-100K, a large-scale paired image-editing dataset with ordered RGBA layer annotations, and introduce dedicated benchmarks for layered editing. Extensive experiments demonstrate that LayerEditor maintains strong editing fidelity while generating accurate RGBA layers, supporting both flexible layer-level manipulation and multi-layer decomposition.