lapisrocks/rpo

Official repository for "Robust Prompt Optimization for Defending Language Models Against Jailbreaking Attacks"

Python

Issues

Adversarial Training that includes PAIR attack
#2 opened 2 months ago by zxzhan
0
About experiments/test.py: Is the [/INST] missing?
#1 opened 8 months ago by xszheng2020
2