Editing Massive Concepts in Text-to-Image Diffusion Models

arXiv:2403.13807v1 Announce Type: cross
Abstract: Text-to-image diffusion models suffer from the risk of generating outdated, copyrighted, incorrect, and biased content. While previous methods have mitigated the issues on a small scale, it is essential to handle them simultaneously in larger-scale real-world scenarios. We propose a two-stage method, Editing Massive Concepts In Diffusion Models (EMCID). The first stage performs memory optimization for each individual concept with dual self-distillation from text alignment loss and diffusion noise prediction loss. The second stage conducts …

abstract arxiv concepts cs.cv cs.lg diffusion diffusion models editing image image diffusion massive risk scale small stage text text-to-image them type world