SCIENCE@home

Shy-hunyuan-MT at WMT25 General Machine Translation Shared Task

Mao Zheng,Zheng Li,Yang Du,Bingxin Qu,Mingyang Song

Published 2025 in Proceedings of the Tenth Conference on Machine Translation

ABSTRACT

In this paper, we present our submission to the WMT25 shared task on machine translation, for which we propose S ynergy-en h anced polic y optimization framework, named Shy . This novel two-phase training framework synergistically combines knowledge distillation and fusion via reinforcement learning. In the first phase, we introduce a multi-stage training framework that harnesses the complementary strengths of multiple state-of-the-art large language models to generate diverse, high-quality translation candidates. These candidates serve as pseudo-references to guide the supervised fine-tuning of our model, Hunyuan-7B, effectively distilling the collective knowledge of multiple expert systems into a single efficient model. In the second phase, we further refine the distilled model through Group Relative Policy Optimization, a reinforcement learning technique that employs a composite reward function. By calculating reward from multiple perspectives, our model ensures better alignment with human preferences and evaluation metrics. Extensive experiments across multiple language pairs demonstrate that our model Shy-hunyuan-MT yields substantial improvements in translation quality compared to baselines. Notably, our framework achieves competitive performance comparable to that of state-of-the-art systems while maintaining computational efficiency through knowledge distillation and fusion.

PUBLICATION RECORD

  • Publication year

    2025

  • Venue

    Proceedings of the Tenth Conference on Machine Translation

  • Publication date

    Unknown publication date

  • Fields of study

    Not labeled

  • Identifiers
    DOI 10.18653/v1/2025.wmt-1.36
  • External record

    Open on Semantic Scholar

  • Source metadata

    Semantic Scholar

CITATION MAP

EXTRACTION MAP

CLAIMS

  • No claims are published for this paper.

CONCEPTS

  • No concepts are published for this paper.

REFERENCES

Showing 1-18 of 18 references · Page 1 of 1

CITED BY