2024-12-04论文泛读《PICCOLO : Exposing Complex Backdoors in NLP Transformer Models》发表时间:2022期刊会议:IEEESymposiumonSecurityandPrivacy(SP)论文单位:PurdueUniversity论文作者:YingqiLiu,GuangyuShen,GuanhongTao,ShengweiAn,ShiqingMa,XiangyuZhang方向分类:BackdoorAttack论文链接开源代码摘要后门可以被注入到NLP模型中,使得当