Cite
Notes
Only stored in your browser.
Attribution
A StrongREJECT for Empty Jailbreaks
arXiv 2024
Refusal-Trained LLMs Are Easily Jailbroken As Browser Agents
from 2 papers
Alexandra Souly
Dillon Bowen
Elaine Chang
Elaine Lau
Elvis Hsieh
Justin Svegliato
Matt Fredrikson
Olivia Watkins
researcher
Pieter Abbeel
professor
Priyanshu Kumar