用检索增强零样本翻译，实现11种东北印度语的高效互译。
BM25-Augmented Many-Shot Translation for Low-Resource North-Eastern Indian Languages
- 基于BM25检索相似双语例句，结合Gemini模型生成翻译。
- 无需微调模型，在22个语言方向上均达到最优性能。
- 适合低资源语言翻译研究者与实际应用开发者参考。
本文介绍佛罗里达大学猎鹰队在WMT26低资源印地语系语言翻译共享任务中的参赛方案。我们将美洲NLP 2026系统中检索增强的多示例翻译流程适配至英语与11种东北印度语之间的双向翻译。推理时，利用BM25从特定语言的训练语料库中检索最相似的平行例句，并由Gemini 2.5 Flash模型在这些例句的条件下生成翻译，全程不涉及模型微调。训练语料库整合了官方的WMT26数据及公开可用语料如Samanantar和历届WMT共享任务发布数据。通过在所有22个语言方向上对检索数量r和开发集示例数d进行网格搜索，为每项提交选择最佳配置。
原文摘要 · Abstract (English)
This paper describes the University of Florida Gators submission to the WMT26 Low-Resource Indic Language Translation shared task. We adapt the retrieval-augmented many-shot translation pipeline from our AmericasNLP 2026 system to translate between English and eleven North-Eastern Indian languages in both directions. At inference time, BM25 retrieves the most similar parallel examples from a language-specific training bank, and Gemini 2.5 Flash translates the input conditioned on these examples. No model fine-tuning is involved. Training banks combine official WMT26 data with publicly available corpora such as Samanantar and prior WMT shared task releases. A grid search over retrieval count r and development exemplar count d across all 22 language-direction pairs selects the best configuration for each submission.
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。