用检索提升低资源语言翻译的少样本学习效果
An Empirical Study of Many-Shot In-Context Learning for Machine Translation of Low-Resource Languages
- 通过BM25检索精选例句，替代大量人工构造示例
- 50个检索例句效果相当于250个原始例句，效率提升4倍
- 对低资源语言翻译有实用价值，适合资源匮乏场景
上下文学习（ICL）使大语言模型能通过少量示例适应新任务，对预训练数据不足的语言具有潜力。近期研究发现，现代大模型可从更长的上下文窗口中获益，但性能依赖于例句选择，且推理成本高。本文针对英语到十种新加入FLORES+的低资源语言的机器翻译任务，实证研究了多示例上下文学习。分析了检索更相关例句、使用跨领域数据以及按长度排序的影响。结果表明，例句数量增加时ICL效果显著提升；更重要的是，基于BM25的检索显著提高数据效率：50个检索例句约等于250个多示例例句的效果，250个检索例句接近1000个原始多示例的效果。此外，ICL在微调基础上仍能带来额外收益。
原文摘要 · Abstract (English)
In-context learning (ICL) allows large language models (LLMs) to adapt to new tasks from a few examples, making it promising for languages underrepresented in pre-training. Recent work on many-shot ICL suggests that modern LLMs can further benefit from larger ICL examples enabled by their long context windows. However, such gains depend on careful example selection, and the inference cost can be prohibitive for low-resource language communities. In this paper, we present an empirical study of many-shot ICL for machine translation from English into ten truly low-resource languages recently added to FLORES+. We analyze the effects of retrieving more informative examples, using out-of-domain data, and ordering examples by length. Our findings show that many-shot ICL becomes more effective as the number of examples increases. More importantly, we show that BM25-based retrieval substantially improves data efficiency: 50 retrieved examples roughly match 250 many-shot examples, while 250 retrieved examples perform similarly to 1,000 many-shot examples. We further show that ICL provides additional gains on top of fine-tuning.
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。