SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding

Zhangchen Xu, Fengqing Jiang, Luyao Niu, Jinyuan Jia, Bill Yuchen Lin, Radha Poovendran

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Fingerprint

Dive into the research topics of 'SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding'. Together they form a unique fingerprint.

Keyphrases

Computer Science