ROSE Doesn't Do That: Boosting the Safety of Instruction-Tuned Large Language Models with Reverse Prompt Contrastive Decoding

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhong, Qihuang, Ding, Liang, Liu, Juhua, Du, Bo, Tao, Dacheng
Format:	Preprint
Published:	2024
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2402.11889