调查多语言RAG系统的语言偏好
Investigating Language Preference of Multilingual RAG Systems
AI总结：
本文系统调查了多语言RAG系统在检索和生成中的语言偏好问题，发现检索器和生成器分别偏向高资源语言和查询语言或拉丁文字导致输出不一致，并提出DKM-RAG框架融合翻译段落与模型知识以缓解该偏好并提升性能。
摘要：
多语言检索增强生成（mRAG）系统通过集成外部多语言信息来增强语言模型，以产生上下文感知的响应。然而，由于查询和文档之间的语言差异，mRAG系统在检索相关信息方面存在困难，当多语言来源发生冲突时会产生不一致的响应。在这项工作中，我们通过一系列实验系统地调查了mRAG在检索和生成中的语言偏好。我们的分析表明，检索器倾向于高资源和查询语言，但这种偏好并不能持续改善生成性能。此外，我们观察到生成器倾向于查询语言或拉丁文字，导致输出不一致。为了克服这些问题，我们提出了双知识多语言RAG（DKM-RAG），这是一个简单而有效的框架，将翻译的多语言段落与互补的模型知识融合。实证结果表明，DKM-RAG减轻了生成中的语言偏好，并增强了各种语言环境下的性能。代码可在https://github.com/jeonghyunpark2002/LanguagePreference.git获取
英文摘要：
Multilingual Retrieval-Augmented Generation (mRAG) systems enhance language models by integrating external multilingual information to produce context-aware responses. However, mRAG systems struggle with retrieving relevant information due to linguistic variations between queries and documents, generating inconsistent responses when multilingual sources conflict. In this work, we systematically investigate language preferences in both retrieval and generation of mRAG through a series of experiments. Our analysis indicates that retrievers tend to prefer high-resource and query languages, yet this preference does not consistently improve generation performance. Moreover, we observe that generators prefer the query language or Latin scripts, leading to inconsistent outputs. To overcome these issues, we propose Dual Knowledge Multilingual RAG (DKM-RAG), a simple yet effective framework that fuses translated multilingual passages with complementary model knowledge. Empirical results demonstrate that DKM-RAG mitigates language preference in generation and enhances performance across diverse linguistic settings. Code is available at https://github.com/jeonghyunpark2002/LanguagePreference.git
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。
