Datasets:

Modalities:
Text
Formats:
arrow
Size:
< 1K
ArXiv:
Libraries:
Datasets
License:
advbench / README.md
fhehli's picture
Update README.md
37d5bf0 verified
metadata
license: mit

Created by Andy Zou and Zifan Wang and J. Zico Kolter and Matt Fredrikson (https://github.com/llm-attacks/llm-attacks)

@misc{zou2023universal,
      title={Universal and Transferable Adversarial Attacks on Aligned Language Models}, 
      author={Andy Zou and Zifan Wang and J. Zico Kolter and Matt Fredrikson},
      year={2023},
      eprint={2307.15043},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}