Text-to-image models suffer from various safety issues that may limit their suitability for deployment. Previous methods have separately addressed individual issues of bias, copyright, and offensive content in text-to-image models. However, in the real world, all of these issues appear simultaneously in the same model. We present a method that tackles all issues with a single approach. Our method, Unified Concept Editing (UCE), edits the model without training using a closed-form solution, and scales seamlessly to concurrent edits on text-conditional diffusion models. We demonstrate scalable simultaneous debiasing, style erasure, and content moderation by editing text-to-image projections, and we present extensive experiments demonstrating improved efficacy and scalability over prior work. Our code is available at https://unified.baulab.info
@article{arxiv.2308.14761,
title = {Unified Concept Editing in Diffusion Models},
author = {Rohit Gandikota and Hadas Orgad and Yonatan Belinkov and Joanna Materzyńska and David Bau},
journal= {arXiv preprint arXiv:2308.14761},
year = {2024}
}
Comments
In proceedings of WACV 2024. Project Page: https://unified.baulab.info