The Art of Defending: A Systematic Evaluation and Analysis of LLM Defense Strategies on Safety and Over-Defensiveness. Varshney, N., Dolin, P., Seth, A., & Baral, C. arXiv preprint arXiv:2401.00287, ACL 2024 (Findings), 2024.
bibtex   
@article{rrv2024chaosw,
  title={The Art of Defending: A Systematic Evaluation and Analysis of LLM Defense Strategies on Safety and Over-Defensiveness},
  author={Varshney, Neeraj and Dolin, Pavel and Seth, Agastya and Baral, Chitta},
  journal={arXiv preprint arXiv:2401.00287},
  publisher="ACL 2024 (Findings)",
  year={2024}
}

Downloads: 0