claws-lab/petgen

A PyTorch implementation of the ACM SIGKDD 2021 paper titled "PETGEN: Personalized Text Generation Attack on Deep Sequence Embedding-based Classification Models"

31
/ 100
Emerging

This project helps online platform moderators and trust & safety teams evaluate the resilience of their deep learning models designed to detect malicious user behavior. It takes a sequence of user posts and a model's existing classification of those posts to generate new, personalized text that malicious users might write to bypass detection. The output is generated text designed to fool the detection model while still appearing plausible.

No commits in the last 6 months.

Use this if you are responsible for maintaining the security and integrity of online platforms and need to proactively test your malicious user detection systems against sophisticated text-based adversarial attacks.

Not ideal if you are looking for a general-purpose text generation tool or a solution to directly filter or block malicious content.

online-trust-and-safety content-moderation fraud-detection user-behavior-analysis platform-security
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 9 / 25

How are scores calculated?

Stars

17

Forks

2

Language

Python

License

MIT

Last pushed

Dec 19, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/claws-lab/petgen"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.