FuxiMT: Sparsifying Large Language Models for Chinese-Centric Multilingual Machine Translation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhu, Shaolin, Dong, Tianyu, Li, Bo, Xiong, Deyi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916746200350720
author Zhu, Shaolin
Dong, Tianyu
Li, Bo
Xiong, Deyi
author_facet Zhu, Shaolin
Dong, Tianyu
Li, Bo
Xiong, Deyi
contents In this paper, we present FuxiMT, a novel Chinese-centric multilingual machine translation model powered by a sparsified large language model (LLM). We adopt a two-stage strategy to train FuxiMT. We first pre-train the model on a massive Chinese corpus and then conduct multilingual fine-tuning on a large parallel dataset encompassing 65 languages. FuxiMT incorporates Mixture-of-Experts (MoEs) and employs a curriculum learning strategy for robust performance across various resource levels. Experimental results demonstrate that FuxiMT significantly outperforms strong baselines, including state-of-the-art LLMs and machine translation models, particularly under low-resource scenarios. Furthermore, FuxiMT exhibits remarkable zero-shot translation capabilities for unseen language pairs, indicating its potential to bridge communication gaps where parallel data are scarce or unavailable.
format Preprint
id arxiv_https___arxiv_org_abs_2505_14256
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FuxiMT: Sparsifying Large Language Models for Chinese-Centric Multilingual Machine Translation
Zhu, Shaolin
Dong, Tianyu
Li, Bo
Xiong, Deyi
Computation and Language
Artificial Intelligence
In this paper, we present FuxiMT, a novel Chinese-centric multilingual machine translation model powered by a sparsified large language model (LLM). We adopt a two-stage strategy to train FuxiMT. We first pre-train the model on a massive Chinese corpus and then conduct multilingual fine-tuning on a large parallel dataset encompassing 65 languages. FuxiMT incorporates Mixture-of-Experts (MoEs) and employs a curriculum learning strategy for robust performance across various resource levels. Experimental results demonstrate that FuxiMT significantly outperforms strong baselines, including state-of-the-art LLMs and machine translation models, particularly under low-resource scenarios. Furthermore, FuxiMT exhibits remarkable zero-shot translation capabilities for unseen language pairs, indicating its potential to bridge communication gaps where parallel data are scarce or unavailable.
title FuxiMT: Sparsifying Large Language Models for Chinese-Centric Multilingual Machine Translation
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2505.14256