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

• URL: http://arxiv.org/abs/2505.14256v1
• Date: Tue, 20 May 2025 12:09:17 GMT
• Title: FuxiMT: Sparsifying Large Language Models for Chinese-Centric Multilingual Machine Translation
• Authors: Shaolin Zhu, Tianyu Dong, Bo Li, Deyi Xiong,
• Abstract summary: We present FuxiMT, a novel Chinese-centric multilingual machine translation model powered by a sparsified large language model (LLM)<n>FuxiMT incorporates Mixture-of-Experts (MoEs) and employs a curriculum learning strategy for robust performance across various resource levels.
• Score: 43.26446958873554
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: 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.

Related papers

• Seed-X: Building Strong Multilingual Translation LLM with 7B Parameters [53.59868121093848]
  We introduce Seed-X, a family of open-source language models (LLMs) with 7B parameter size.<n>The base model is pre-trained on a diverse, high-quality dataset encompassing both monolingual and bilingual content across 28 languages.<n>The instruct model is then finetuned to translate by Chain-of-Thought (CoT) reasoning and further enhanced through reinforcement learning (RL) to achieve better generalization across diverse language pairs.
  arXiv  Detail & Related papers  (2025-07-18T03:19:43Z)
• Trans-Zero: Self-Play Incentivizes Large Language Models for Multilingual Translation Without Parallel Data [64.4458540273004]
  We propose a self-play framework that leverages only monolingual data and the intrinsic multilingual knowledge of Large Language Models (LLMs)<n>Experiments demonstrate that this approach not only matches the performance of models trained on large-scale parallel data but also excels in non-English translation directions.
  arXiv  Detail & Related papers  (2025-04-20T16:20:30Z)
• Multilingual Machine Translation with Open Large Language Models at Practical Scale: An Empirical Study [13.409987421121405]
  GemmaX2-28 is a 9B model achieving top-tier multilingual translation performance across 28 languages.<n>GemmaX2-28 consistently outperforms the state-of-the-art (SOTA) models such as TowerInstruct and XALMA.
  arXiv  Detail & Related papers  (2025-02-04T16:57:03Z)
• Quality or Quantity? On Data Scale and Diversity in Adapting Large Language Models for Low-Resource Translation [62.202893186343935]
  We explore what it would take to adapt Large Language Models for low-resource languages. We show that parallel data is critical during both pre-training andSupervised Fine-Tuning (SFT) Our experiments with three LLMs across two low-resourced language groups reveal consistent trends, underscoring the generalizability of our findings.
  arXiv  Detail & Related papers  (2024-08-23T00:59:38Z)
• FuxiTranyu: A Multilingual Large Language Model Trained with Balanced Data [39.54285525397304]
  We present FuxiTranyu, an open-source multilingual model for large language models (LLMs) The base model, FuxiTranyu-8B, features 8 billion parameters and is trained from scratch on meticulously balanced multilingual data. Experiments on a wide range of multilingual benchmarks demonstrate the competitive performance of FuxiTranyu.
  arXiv  Detail & Related papers  (2024-08-12T16:34:56Z)
• To Translate or Not to Translate: A Systematic Investigation of Translation-Based Cross-Lingual Transfer to Low-Resource Languages [0.0]
  We evaluate existing and propose new translation-based XLT approaches for transfer to low-resource languages. We show that all translation-based approaches dramatically outperform zero-shot XLT with mLMs. We propose an effective translation-based XLT strategy even for languages not supported by the MT system.
  arXiv  Detail & Related papers  (2023-11-15T22:03:28Z)
• Building Multilingual Machine Translation Systems That Serve Arbitrary X-Y Translations [75.73028056136778]
  We show how to practically build MNMT systems that serve arbitrary X-Y translation directions. We also examine our proposed approach in an extremely large-scale data setting to accommodate practical deployment scenarios.
  arXiv  Detail & Related papers  (2022-06-30T02:18:15Z)
• Towards the Next 1000 Languages in Multilingual Machine Translation: Exploring the Synergy Between Supervised and Self-Supervised Learning [48.15259834021655]
  We present a pragmatic approach towards building a multilingual machine translation model that covers hundreds of languages. We use a mixture of supervised and self-supervised objectives, depending on the data availability for different language pairs. We demonstrate that the synergy between these two training paradigms enables the model to produce high-quality translations in the zero-resource setting.
  arXiv  Detail & Related papers  (2022-01-09T23:36:44Z)
• Pre-training Multilingual Neural Machine Translation by Leveraging Alignment Information [72.2412707779571]
  mRASP is an approach to pre-train a universal multilingual neural machine translation model. We carry out experiments on 42 translation directions across a diverse setting, including low, medium, rich resource, and as well as transferring to exotic language pairs.
  arXiv  Detail & Related papers  (2020-10-07T03:57:54Z)
• Improving Massively Multilingual Neural Machine Translation and Zero-Shot Translation [81.7786241489002]
  Massively multilingual models for neural machine translation (NMT) are theoretically attractive, but often underperform bilingual models and deliver poor zero-shot translations. We argue that multilingual NMT requires stronger modeling capacity to support language pairs with varying typological characteristics. We propose random online backtranslation to enforce the translation of unseen training language pairs.
  arXiv  Detail & Related papers  (2020-04-24T17:21:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.