用检索增强生成修复极低资源语言翻译的领域漂移问题
Context Volume Drives Performance: Tackling Domain Shift in Extremely Low-Resource Translation via RAG
- 先用NMT生成初稿，再由LLM结合检索内容精修
- 在未见领域下将翻译得分从27.11提升至35.21
- 检索条目数量比算法选择更影响最终效果
针对极低资源语言的神经机器翻译模型在遭遇领域漂移时性能显著下降。我们以印度东部分布的土著语言Dhao为例，该语言仅存新约圣经（NT）的数字化文本。当将训练于新约的模型用于未见的旧约（OT）时，其chrF++得分从域内36.17降至27.11。为恢复性能，我们提出混合框架：先用微调后的NMT生成初稿，再通过大语言模型（LLM）结合检索增强生成（RAG）进行优化。最终系统达35.21 chrF++，实现8.10的性能回升，基本复现原域内水平。分析表明，性能提升主要依赖于检索到的示例数量，而非检索算法本身。定性分析显示，LLM充当稳健的“安全网”，可修复零样本场景下的严重错误。
原文摘要 · Abstract (English)
Neural Machine Translation (NMT) models for low-resource languages suffer significant performance degradation under domain shift. We quantify this challenge using Dhao, an indigenous language of Eastern Indonesia with no digital footprint beyond the New Testament (NT). When applied to the unseen Old Testament (OT), a standard NMT model fine-tuned on the NT drops from an in-domain score of 36.17 chrF++ to 27.11 chrF++. To recover this loss, we introduce a hybrid framework where a fine-tuned NMT model generates an initial draft, which is then refined by a Large Language Model (LLM) using Retrieval-Augmented Generation (RAG). The final system achieves 35.21 chrF++ (+8.10 recovery), effectively matching the original in-domain quality. Our analysis reveals that this performance is driven primarily by the number of retrieved examples rather than the choice of retrieval algorithm. Qualitative analysis confirms the LLM acts as a robust "safety net," repairing severe failures in zero-shot domains.
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。