无需双语数据，大模型自玩自纠实现多语言翻译
Trans-Zero: Self-Play Incentivizes Large Language Models for Multilingual Translation Without Parallel Data
- 用自对弈机制在单语数据中迭代优化翻译
- 在非英语方向上超越传统方法，媲美有监督训练
- 适合低资源语言翻译与避免遗忘的场景
大语言模型（LLM）虽重塑了机器翻译，但多语言翻译仍依赖平行数据进行有监督微调，面临低资源语言数据稀缺和灾难性遗忘等问题。为此，我们提出TRANS-ZERO，一种仅使用单语数据和LLM固有多语言知识的自对弈框架。该框架结合遗传蒙特卡洛树搜索（G-MCTS）与偏好优化，在无平行数据条件下实现了媲美有监督方法的翻译性能。实验表明，该方法不仅在大规模平行数据训练模型的性能上持平，更在非英语翻译方向表现优异。进一步分析显示，G-MCTS通过迭代翻译探索语义一致候选，显著提升翻译质量，为框架成功提供坚实基础。
原文摘要 · Abstract (English)
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 succuss.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。