LauJames/Topic-FlipRAG
[USENIX Security 2025] Topic-FlipRAG: Topic-Orientated Adversarial Opinion Manipulation Attacks to Retrieval-Augmented Generation Models
This project helps AI safety researchers and red teamers evaluate the robustness of Retrieval-Augmented Generation (RAG) systems. It takes an opinion dataset and, through a process of generating 'poisoned' documents, demonstrates how to subtly manipulate the RAG system's output to reflect a specific opinion across related queries. The end-user is typically an AI security expert or a researcher focusing on adversarial machine learning.
No commits in the last 6 months.
Use this if you need to test how easily a RAG model's opinion stance can be flipped by introducing a small number of carefully crafted, adversarial documents into its knowledge base.
Not ideal if you are looking to build or enhance a RAG system for production use, as this tool is specifically designed for adversarial testing and manipulation.
Stars
10
Forks
2
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Jun 21, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/rag/LauJames/Topic-FlipRAG"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
LLAMATOR-Core/llamator
Red Teaming python-framework for testing chatbots and GenAI systems.
sleeepeer/PoisonedRAG
[USENIX Security 2025] PoisonedRAG: Knowledge Corruption Attacks to Retrieval-Augmented...
kelkalot/simpleaudit
Allows to red-team your AI systems through adversarial probing. It is simple, effective, and...
JuliusHenke/autopentest
CLI enabling more autonomous black-box penetration tests using Large Language Models (LLMs)
SecurityClaw/SecurityClaw
A modular, skill-based autonomous Security Operations Center (SOC) agent that monitors...