Saved in:
Bibliographic Details
Main Authors: Khader, Massine El, Bouzidi, Elias Al, Oumida, Abdellah, Sbaihi, Mohammed, Binard, Eliott, Poli, Jean-Philippe, Ouerdane, Wassila, Addad, Boussad, Kapusta, Katarzyna
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.00064
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912237177798656
author Khader, Massine El
Bouzidi, Elias Al
Oumida, Abdellah
Sbaihi, Mohammed
Binard, Eliott
Poli, Jean-Philippe
Ouerdane, Wassila
Addad, Boussad
Kapusta, Katarzyna
author_facet Khader, Massine El
Bouzidi, Elias Al
Oumida, Abdellah
Sbaihi, Mohammed
Binard, Eliott
Poli, Jean-Philippe
Ouerdane, Wassila
Addad, Boussad
Kapusta, Katarzyna
contents Recent advances in Diffusion Models have enabled the generation of images from text, with powerful closed-source models like DALL-E and Midjourney leading the way. However, open-source alternatives, such as StabilityAI's Stable Diffusion, offer comparable capabilities. These open-source models, hosted on Hugging Face, come equipped with ethical filter protections designed to prevent the generation of explicit images. This paper reveals first their limitations and then presents a novel text-based safety filter that outperforms existing solutions. Our research is driven by the critical need to address the misuse of AI-generated content, especially in the context of information warfare. DiffGuard enhances filtering efficacy, achieving a performance that surpasses the best existing filters by over 14%.
format Preprint
id arxiv_https___arxiv_org_abs_2412_00064
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DiffGuard: Text-Based Safety Checker for Diffusion Models
Khader, Massine El
Bouzidi, Elias Al
Oumida, Abdellah
Sbaihi, Mohammed
Binard, Eliott
Poli, Jean-Philippe
Ouerdane, Wassila
Addad, Boussad
Kapusta, Katarzyna
Computer Vision and Pattern Recognition
Artificial Intelligence
Recent advances in Diffusion Models have enabled the generation of images from text, with powerful closed-source models like DALL-E and Midjourney leading the way. However, open-source alternatives, such as StabilityAI's Stable Diffusion, offer comparable capabilities. These open-source models, hosted on Hugging Face, come equipped with ethical filter protections designed to prevent the generation of explicit images. This paper reveals first their limitations and then presents a novel text-based safety filter that outperforms existing solutions. Our research is driven by the critical need to address the misuse of AI-generated content, especially in the context of information warfare. DiffGuard enhances filtering efficacy, achieving a performance that surpasses the best existing filters by over 14%.
title DiffGuard: Text-Based Safety Checker for Diffusion Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2412.00064