作者位次 #5/6ACL 2026Poster· Natural Language Processing
M irage B ackdoor: A Stealthy Attack that Induces Think-Well-Answer-Wrong Reasoning
Yizhe Zeng · Wei Zhang · Yunpeng Li · Juxin Xiao · Xiao Wang · Yuling Liu
摘要
While Chain-of-Thought (CoT) prompting has become a standard paradigm for eliciting complex reasoning capabilities in Large Language Models, it inadvertently exposes a new attack surface for backdoor attacks. Existing CoT backdoor attacks typically manipulate the intermediate reasoning steps to s...