破解多语言RAG系统语言偏好难题，提升跨语言生成一致性
Investigating Language Preference of Multilingual RAG Systems
- 通过实验发现检索器偏爱高资源语种和查询语种
- 生成器倾向使用查询语种或拉丁字母，导致输出不一致
- 提出DKM-RAG框架融合翻译文本与模型知识，有效缓解语言偏好
多语言检索增强生成（mRAG）系统通过整合外部多语言信息来生成上下文感知的回答。然而，由于查询与文档间存在语言差异，mRAG系统在检索相关性时表现不佳，当多语言来源冲突时会产生不一致的响应。本文通过一系列实验系统研究了mRAG在检索与生成阶段的语言偏好。分析表明，检索器倾向于偏好高资源语言和查询语言，但这种偏好并未持续提升生成性能。此外，生成器更倾向于使用查询语言或拉丁字符，导致输出不一致。为此，我们提出双重知识多语言RAG（DKM-RAG），一种简单而有效的框架，通过融合翻译后的多语言段落与互补的模型知识来克服这些问题。实证结果表明，DKM-RAG有效缓解了生成阶段的语言偏好，并在多种语言环境下提升了性能。代码已开源：https://github.com/jeonghyunpark2002/LanguagePreference.git
原文摘要 · Abstract (English)
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
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。