AI RESEARCH
Masked Generative Transformer Is What You Need for Image Editing
arXiv CS.LG
•
ArXi:2605.10859v1 Announce Type: cross Diffusion models dominate image editing, yet their global denoising mechanism entangles edited regions with surrounding context, causing modifications to propagate into areas that should remain intact. We propose a fundamentally different approach by leveraging Masked Generative Transformers (MGTs), whose localized token-prediction paradigm naturally confines changes to intended regions. We present EditMGT, an MGT-based editing framework that is the first of its kind.