ML Anthology

ChatBug: A Common Vulnerability of Aligned LLMs Induced by Chat Templates

Fengqing Jiang, Zhangchen Xu, Luyao Niu, Bill Yuchen Lin, Radha Poovendran
AAAI 2025 pp. 27347-27355
doi:10.1609/AAAI.V39I26.34945 /aaai/2025/jiang2025aaai-chatbug/
AAAI Semantic Scholar

Cite

Text

Jiang et al. "ChatBug: A Common Vulnerability of Aligned LLMs Induced by Chat Templates." AAAI Conference on Artificial Intelligence, 2025. doi:10.1609/AAAI.V39I26.34945

Markdown

[Jiang et al. "ChatBug: A Common Vulnerability of Aligned LLMs Induced by Chat Templates." AAAI Conference on Artificial Intelligence, 2025.](https://mlanthology.org/aaai/2025/jiang2025aaai-chatbug/) doi:10.1609/AAAI.V39I26.34945

BibTeX

@inproceedings{jiang2025aaai-chatbug,
  title     = {{ChatBug: A Common Vulnerability of Aligned LLMs Induced by Chat Templates}},
  author    = {Jiang, Fengqing and Xu, Zhangchen and Niu, Luyao and Lin, Bill Yuchen and Poovendran, Radha},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2025},
  pages     = {27347-27355},
  doi       = {10.1609/AAAI.V39I26.34945},
  url       = {https://mlanthology.org/aaai/2025/jiang2025aaai-chatbug/}
}