JailPO: A Novel Black-Box Jailbreak Framework via Preference Optimization Against Aligned LLMs

Cite

Text

Li et al. "JailPO: A Novel Black-Box Jailbreak Framework via Preference Optimization Against Aligned LLMs." AAAI Conference on Artificial Intelligence, 2025. doi:10.1609/AAAI.V39I26.34953

Markdown

[Li et al. "JailPO: A Novel Black-Box Jailbreak Framework via Preference Optimization Against Aligned LLMs." AAAI Conference on Artificial Intelligence, 2025.](https://mlanthology.org/aaai/2025/li2025aaai-jailpo/) doi:10.1609/AAAI.V39I26.34953

BibTeX

@inproceedings{li2025aaai-jailpo,
  title     = {{JailPO: A Novel Black-Box Jailbreak Framework via Preference Optimization Against Aligned LLMs}},
  author    = {Li, Hongyi and Ye, Jiawei and Wu, Jie and Yan, Tianjie and Wang, Chu and Li, Zhixin},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2025},
  pages     = {27419-27427},
  doi       = {10.1609/AAAI.V39I26.34953},
  url       = {https://mlanthology.org/aaai/2025/li2025aaai-jailpo/}
}