AUTODAN:生成对齐大型语言模型上的隐形越狱提示.docx

AUTODAN:生成对齐大型语言模型上的隐形越狱提示.docx

PublishedasaconferencepaperatICLR2024

1

AUTODAN:GENERATINGSTEALTHYJAILBREAKPROMPTSONALIGNEDLARGELANGUAGEMODELS

XiaogengLiu1NanXu2MuhaoChen3ChaoweiXiao1

1UniversityofWisconsin–Madison,2USC,3UniversityofCalifornia,Davis

ABSTRACT

arXi

文档评论(0)

1亿VIP精品文档

相关文档