Backdoors Against Natural Language Processing: A Review

Shaofeng Li, Tian Dong, Benjamin Zi Hao Zhao, Minhui Xue, Suguo Du, Haojin Zhu

Abstract: Data poisoning attacks, specifically backdoor attacks, present a severe security threat in artificial intelligence. We provide a comprehensive survey into state-of-the-art backdoor attacks and defenses in the field of natural language processing.