Soft Begging: Modular and Efficient Shielding of LLMs against Prompt Injection and Jailbreaking based on Prompt Tuning. Ostermann, S., Baum, K., Endres, C., Masloh, J., & Schramowski, P. 2024.
Soft Begging: Modular and Efficient Shielding of LLMs against Prompt Injection and Jailbreaking based on Prompt Tuning [link]Paper  bibtex   1 download  
@misc{ostermann2024softbeggingmodularefficient,
      title={Soft Begging: Modular and Efficient Shielding of LLMs against Prompt Injection and Jailbreaking based on Prompt Tuning}, 
      author={Simon Ostermann and Kevin Baum and Christoph Endres and Julia Masloh and Patrick Schramowski},
      year={2024},
      eprint={2407.03391},
      archivePrefix={arXiv},
      primaryClass={cs.CR},
      url={https://arxiv.org/abs/2407.03391}, 
}

Downloads: 1