AI RESEARCH

Semantic Granularity Navigation in Image Editing

arXiv CS.CV

ArXi:2605.21190v1 Announce Type: new Despite the generative capabilities of diffusion and flow models, real-image editing remains constrained by a persistent trade-off between semantic editability and structural fidelity. We trace a primary cause of this limitation to the implicit coupling of edit progress with model scale in existing paradigms. Under this coupling, stronger edits typically require visiting noisier states, which spends computation on destabilizing layout before the semantic change is well localized. We.