Divide-and-Conquer Attack

This is the official implementation for paper: Divide-and-Conquer Attack: Harnessing the Power of LLM to Bypass the Censorship of Text-to-Image Generation Model

Warning: This repository may contain harmful content.

Dataset

The dataset Sensitive Prompt Dataset.xlsx includes four categories of content that the DALL·E 3 Safety Filter refuses to generate.

Gradio Demo:

• Download the source file DACADemo.py and the requirements file requirements.txt
• Install the required libraries: pip install -r requirements.txt
• Run the source file: python DACADemo.py

Gallery

The gallery contains some representative images generated by DALL·E 3 during the evaluation process of the paper.

Demonstration Video

The video demonstrates the use of the Gradio demo to execute DACA, including three parts: Copyright Character, Discriminatory Content, and Re-use Adversarial Prompt.