CLAIJul 1, 2024

Retrieval-augmented generation in multilingual settings

arXiv:2407.01463v140 citationsh-index: 26Has Code
Originality Synthesis-oriented
AI Analysis

This work addresses the gap in RAG research for non-English languages, providing a baseline for future multilingual applications, though it is incremental as it adapts existing methods to new data.

The paper tackled the problem of applying retrieval-augmented generation (RAG) to multilingual settings, finding that task-specific prompt engineering is needed for generation in user languages and that current evaluation metrics require adjustments for multilingual variations.

Retrieval-augmented generation (RAG) has recently emerged as a promising solution for incorporating up-to-date or domain-specific knowledge into large language models (LLMs) and improving LLM factuality, but is predominantly studied in English-only settings. In this work, we consider RAG in the multilingual setting (mRAG), i.e. with user queries and the datastore in 13 languages, and investigate which components and with which adjustments are needed to build a well-performing mRAG pipeline, that can be used as a strong baseline in future works. Our findings highlight that despite the availability of high-quality off-the-shelf multilingual retrievers and generators, task-specific prompt engineering is needed to enable generation in user languages. Moreover, current evaluation metrics need adjustments for multilingual setting, to account for variations in spelling named entities. The main limitations to be addressed in future works include frequent code-switching in non-Latin alphabet languages, occasional fluency errors, wrong reading of the provided documents, or irrelevant retrieval. We release the code for the resulting mRAG baseline pipeline at https://github.com/naver/bergen.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes