@inproceedings{zou-etal-2025-trans,
    title = "{TRANS}-{ZERO}: Self-Play Incentivizes Large Language Models for Multilingual Translation Without Parallel Data",
    author = "Zou, Wei  and
      Yang, Sen  and
      Bao, Yu  and
      Huang, Shujian  and
      Chen, Jiajun  and
      Cheng, Shanbo",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingestion-acl-25/2025.findings-acl.637/",
    pages = "12337--12347",
    ISBN = "979-8-89176-256-5",
    abstract = "The rise of Large Language Models (LLMs) has reshaped machine translation (MT), but multilingual MT still relies heavily on parallel data for supervised fine-tuning (SFT), facing challenges like data scarcity for low-resource languages and catastrophic forgetting. To address these issues, we propose TRANS-ZERO, a self-play framework that leverages only monolingual data and the intrinsic multilingual knowledge of LLM. TRANS-ZERO combines Genetic Monte-Carlo Tree Search (G-MCTS) with preference optimization, achieving strong translation performance that rivals supervised methods. 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. Further analysis reveals that G-MCTS itself significantly enhances translation quality by exploring semantically consistent candidates through iterative translations, providing a robust foundation for the framework{'}s success."
}