Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Training
Youliang Yuan, Wenxiang Jiao, Wenxuan Wang, Jen-tse Huang, Jiahao Xu, Tian Liang, Pinjia He, Zhaopeng Tu. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.
Authors
- Wenxiang Jiao (ORCID: https://orcid.org/0000-0003-4951-9420)
- Liang Tian (ORCID: https://orcid.org/0000-0002-6595-445X)
- Jen-tse Huang (ORCID: https://orcid.org/0000-0003-3446-0083)
- Wenxuan Wang (ORCID: https://orcid.org/0000-0001-7172-8662)
- Youliang Yuan
- Jiahao Xu (ORCID: https://orcid.org/0000-0001-9894-228X)
- Pinjia He (ORCID: https://orcid.org/0000-0003-3377-8129)
- Zhaopeng Tu (ORCID: https://orcid.org/0000-0002-7900-6055)
Publication Details
- Published
- 2025-01-01
- DOI
- https://doi.org/10.18653/v1/2025.acl-long.158
- Citations
- 1
- Primary Topic
- Safety Systems Engineering in Autonomy
- Type
- article
- Field-Weighted Citation Impact
- 1.30
Funders
- Basic and Applied Basic Research Foundation of Guangdong Province