An Optimizable Suffix Is Worth A Thousand Templates: Efficient Black-box Jailbreaking without Affirmative Phrases via LLM as Optimizer

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Jiang, Weipeng, Wang, Zhenting, Zhai, Juan, Ma, Shiqing, Zhao, Zhengyu, Shen, Chao
Format:	Preprint
Published:	2024
Subjects:	Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2408.11313