Found 2 repositories(showing 2)
facebookresearch
Repo for the paper "Meta SecAlign: A Secure Foundation LLM Against Prompt Injection Attacks".
sanchayan7432
Secure LLM Align is a defending method against PLeak attacks at an advanced level. The concept is collected from the paper SecAlign, research by Meta. Link : https://arxiv.org/abs/2410.05451
All 2 repositories loaded