• search
    search
  • reviewers
    reviewers
  • feeds
    feeds
  • assignments
    assignments
  • settings
  • logout

The Art of Defending: A Systematic Evaluation and Analysis of LLM Defense Strategies on Safety and Over-Defensiveness.

Neeraj VarshneyPavel DolinAgastya SethChitta Baral
Published in: CoRR (2024)
Keyphrases
  • systematic evaluation
  • data analysis
  • artificial intelligence
  • feature space
  • language model
  • statistical analysis
  • high precision