No One-Size-Fits-All: Building Systems For Translation to Bashkir, Kazakh, Kyrgyz, Tatar and Chuvash Using Synthetic And Original Data
arXiv: 2602.04442v1
发布: 2026-02-04
更新: 2026-02-04
AI 摘要
该论文研究了五种突厥语机器翻译，利用合成数据和检索方法优化了翻译效果。
主要贡献
- 针对五种突厥语的机器翻译模型构建
- 利用合成数据微调模型，提升翻译效果
- 使用检索方法辅助翻译
- 发布数据集和模型权重
方法论
该论文使用了LoRA微调、Prompting DeepSeek-V3.2、零样本学习和检索等方法进行机器翻译。
原文摘要
We explore machine translation for five Turkic language pairs: Russian-Bashkir, Russian-Kazakh, Russian-Kyrgyz, English-Tatar, English-Chuvash. Fine-tuning nllb-200-distilled-600M with LoRA on synthetic data achieved chrF++ 49.71 for Kazakh and 46.94 for Bashkir. Prompting DeepSeek-V3.2 with retrieved similar examples achieved chrF++ 39.47 for Chuvash. For Tatar, zero-shot or retrieval-based approaches achieved chrF++ 41.6, while for Kyrgyz the zero-shot approach reached 45.6. We release the dataset and the obtained weights.