MPO: Multilingual Safety Alignment via Reward Gap Optimization
Weixiang Zhao, Yulin Hu, Yang Deng, Tongtong Wu, Wenxuan Zhang, Jiahe Guo, An Zhang, Yanyan Zhao
Abstract
Large language models (LLMs) have become increasingly central to AI applications worldwide, necessitating robust multilingual safety alignment to ensure secure deployment across diverse linguistic contexts. Existing preference learning methods for safety alignment, such as RLHF and DPO, are primarily monolingual and struggle with noisy multilingual data. To address these limitations, we introduce Multilingual reward gaP Optimization (MPO), a novel approach that leverages the well-aligned safety capabilities of the dominant language (e.g., English) to improve safety alignment across multiple languages. MPO directly minimizes the reward gap difference between the dominant language and target languages, effectively transferring safety capabilities while preserving the original strengths of the dominant language. Extensive experiments on three LLMs, LLaMA-3.1, Gemma-2 and Qwen2.5, validate MPO’s efficacy in multilingual safety alignment without degrading general multilingual utility.
BibTeX
@inproceedings{zhao-etal-2025-mpo,
title = "{MPO}: Multilingual Safety Alignment via Reward Gap Optimization",
author = "Zhao, Weixiang and
Hu, Yulin and
Deng, Yang and
Wu, Tongtong and
Zhang, Wenxuan and
Guo, Jiahe and
Zhang, An and
Zhao, Yanyan and
Qin, Bing and
Chua, Tat-Seng and
Liu, Ting",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.acl-long.1149/",
doi = "10.18653/v1/2025.acl-long.1149",
pages = "23564--23587",
ISBN = "979-8-89176-251-0"
}