Login / Signup

ChatGPT as an Attack Tool: Stealthy Textual Backdoor Attack via Blackbox Generative Model Trigger.

Jiazhao LiYijin YangZhuofeng WuV. G. Vinod VydiswaranChaowei Xiao
Published in: NAACL-HLT (2024)
Keyphrases
  • generative model
  • probabilistic model
  • em algorithm
  • bayesian framework
  • semi supervised
  • mixture model
  • posterior probability
  • discriminative learning
  • keywords
  • training set
  • topic models
  • object categories