@inproceedings{zhu-etal-2024-flipguard,
title = "{F}lip{G}uard: Defending Preference Alignment against Update Regression with Constrained Optimization",
author = "Zhu, Mingye and
Liu, Yi and
Wang, Quan and
Guo, Junbo and
Mao, Zhendong",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://preview.aclanthology.org/Add-Cong-Liu-Florida-Atlantic-University-author-id/2024.emnlp-main.960/",
doi = "10.18653/v1/2024.emnlp-main.960",
pages = "17333--17350"
}
Markdown (Informal)
[FlipGuard: Defending Preference Alignment against Update Regression with Constrained Optimization](https://preview.aclanthology.org/Add-Cong-Liu-Florida-Atlantic-University-author-id/2024.emnlp-main.960/) (Zhu et al., EMNLP 2024)
ACL