Chain-of-Scrutiny: Detecting Backdoor Attacks for Large Language Models.

Xi Li, Ruofan Mao, Yusen Zhang, Renze Lou, Chen Wu, Jiaqi Wang

Browse the full ACL paper archive.