SurrogatePrompt: Bypassing the Safety Filter of Text-to-Image Models via Substitution

Image Censorship

XCloud. 2019 · 2019

Earlier work this paper cites.

ArcFace: Additive Angular Margin Loss for Deep Face Recognition

Jiankang Deng, Jia Guo, Jing Yang, Niannan Xue, Irene Kotsia, and Stefanos Zafeiriou. 2022 · 2021

Earlier work this paper cites.

Discovering the Hidden Vocabulary of DALLE-2

Original

Giannis Daras and Alexandros G. Dimakis. 2022 · 2022

Earlier work this paper cites.

Unstable Diffusion: Ethical challenges and some ways forward

Abhishek Gupta. 2022 · 2022

Earlier work this paper cites.

CLIP-based-NSFW-Detector

LAION-AI. 2022 · 2022

Earlier work this paper cites.

Adversarial attacks on image generation with made-up words

Original

Raphaël Millière. 2022 · 2022

Earlier work this paper cites.

Hierarchical Text-Conditional Image Generation with CLIP Latents

Original

Aditya Ramesh, Prafulla Dhariwal, Alex Nichol, Casey Chu, and Mark Chen. 2022 · 2022

Earlier work this paper cites.

Red-teaming the stable diffusion safety filter

Original

Javier Rando, Daniel Paleka, David Lindner, Lennard Heim, and Florian Tramèr. 2022 · 2022

Earlier work this paper cites.

High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition . 10684–10695

Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. 2022 · 2022

Earlier work this paper cites.

Can Machines Help Us Answering Question 16 in Datasheets, and In Turn Reflecting on Inappropriate Content?

Original

Patrick Schramowski, Christopher Tauchmann, and Kristian Kersting. 2022 · 2022

Earlier work this paper cites.

Midjourney now has an easy mode and a new AI moderation system

Matthias Bastian. 2023 · 2023

Cited alongside, same era.

Extracting training data from diffusion models

Original

Nicholas Carlini, Jamie Hayes, Milad Nasr, Matthew Jagielski, Vikash Sehwag, Florian Tramer, Borja Balle, Daphne Ippolito, and Eric Wallace. 2023 · 2023

Cited alongside, same era.

The Hidden Language of Diffusion Models

Original

Hila Chefer, Oran Lang, Mor Geva, Volodymyr Polosukhin, Assaf Shocher, Michal Irani, Inbar Mosseri, and Lior Wolf. 2023 · 2023

Cited alongside, same era.

Divide-and-Conquer Attack: Harnessing the Power of LLM to Bypass the Censorship of Text-to-Image Generation Model

Original

Yimo Deng and Huangxun Chen. 2023 · 2023

Cited alongside, same era.

Terms of Service

DreamStudio. 2023 · 2023

Cited alongside, same era.

Unsafe Diffusion: On the Generation of Unsafe Images and Hateful Memes From Text-To-Image Models

Original

Yiting Qu, Xinyue Shen, Xinlei He, Michael Backes, Savvas Zannettou, and Yang Zhang. 2023 · 2023

Closest in time.

Midjourney bans a user who generates images of politicians cheating on their wives

Walker Ronnie. 2023 · 2023

Closest in time.

Safe latent diffusion: Mitigating inappropriate degeneration in diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 22522–22531

Patrick Schramowski, Manuel Brack, Björn Deiseroth, and Kristian Kersting. 2023 · 2023

Closest in time.

A Reproducible Extraction of Training Images from Diffusion Models

Original

Ryan Webster. 2023 · 2023

Closest in time.

Midjourney Statistics: Users, Polls, & Growth [Aug 2023]

Andrew Wilson. 2023 · 2023