Analyzing and Editing Inner Mechanisms of Backdoored Language Models.
PROCEEDINGS OF THE 2024 ACM CONFERENCE ON FAIRNESS, ACCOUNTABILITY, AND TRANSPARENCY, ACM FACCT 2024（2024）
关键词
Interpretability,Backdoor Attacks,Backdoor Defenses,Natural Language Processing,Safety
AI 理解论文
溯源树
样例
生成溯源树，研究论文发展脉络
Chat Paper
正在生成论文摘要