基于非结构化知识的检索增强机器翻译
Retrieval-Augmented Machine Translation with Unstructured Knowledge
AI总结：
针对非结构化、跨语言不完全配对的世界知识，本文构建含169K样本的RAGtrans基准，并提出无需额外标注的多任务训练方法，显著提升LLM多语言检索增强翻译性能。
摘要：
检索增强生成（RAG）通过引入额外信息来增强大型语言模型（LLM）。在机器翻译（MT）中，以往工作通常从配对的机器翻译语料库中检索上下文示例，或从知识图谱中检索领域特定知识，以增强机器翻译模型。然而，大量世界知识以非结构化文档形式组织，并且在不同语言之间可能并非完全配对。本文研究利用非结构化文档进行检索增强机器翻译。具体而言，我们构建了 RAGtrans，这是首个用于训练和评估 LLM 检索增强机器翻译能力的基准。RAGtrans 包含通过 GPT-4o 和人工译员收集的 169K 个机器翻译样本。此外，还提供来自多种语言的文档，为这些样本补充知识。基于 RAGtrans，我们进一步提出一种多任务训练方法，教 LLM 在翻译过程中如何利用多语言文档中的信息。该方法利用现有多语言语料库创建辅助训练目标，无需额外标注。大量实验表明，该方法在英译汉中使 LLM 提升 1.6–3.1 BLEU 和 1.0–2.0 COMET 分数，在英译德中提升 1.7–2.9 BLEU 和 2.1–2.7 COMET 分数。我们还总结了当前 LLM 在该任务上面临的关键困难。
英文摘要：
Retrieval-augmented generation (RAG) introduces additional information to enhance large language models (LLMs). In machine translation (MT), previous work typically retrieves in-context examples from paired MT corpora, or domain-specific knowledge from knowledge graphs, to enhance MT models. However, a large amount of world knowledge is organized in unstructured documents, and might not be fully paired across different languages. In this paper, we study retrieval-augmented MT using unstructured documents. Specifically, we build RAGtrans, the first benchmark to train and evaluate LLMs' retrieval-augmented MT ability. RAGtrans contains 169K MT samples collected via GPT-4o and human translators. Besides, documents from various languages are also provided to supply the knowledge to these samples. Based on RAGtrans, we further propose a multi-task training method to teach LLMs how to use information from multilingual documents during their translation. The method uses existing multilingual corpora to create auxiliary training objectives without additional labeling requirements. Extensive experiments show that the method improves LLMs by 1.6-3.1 BLEU and 1.0-2.0 COMET scores in En-Zh, and 1.7-2.9 BLEU and 2.1-2.7 COMET scores in En-De. We also conclude the critical difficulties that current LLMs face with this task.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
