Login / Signup

Red Alarm for Pre-trained Models: Universal Vulnerability to Neuron-level Backdoor Attacks.

Zhengyan ZhangGuangxuan XiaoYongwei LiTian LvFanchao QiZhiyuan LiuYasheng WangXin JiangMaosong Sun
Published in: Mach. Intell. Res. (2023)
Keyphrases
  • pre trained
  • data sets
  • neural network
  • computer vision