# All Languages Matter: Understanding and Mitigating Language Bias in Multilingual RAG

Source: [All Languages Matter: Understanding and Mitigating Language Bias in Multilingual RAG](<https://www.sota2.com/research/paper/69e9a7ad55a3c32af89f717c>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-24T05:00:04.605Z

Authors: Dan Wang, Guozhao Mo, Yafei Shi, Cheng Zhang, Bo Zheng, Boxi Cao, Xuanang Chen, Yaojie Lu, Hongyu Lin, Ben He, Xianpei Han, Le Sun

Published: 2026-04-22T05:33:06.000Z

## Abstract

Multilingual Retrieval-Augmented Generation (mRAG) leverages cross-lingual evidence to ground Large Language Models (LLMs) in global knowledge. However, we show that current mRAG systems suffer from a language bias during reranking, systematically favoring English and the query's native language. By introducing an estimated oracle evidence analysis, we quantify a substantial performance gap between existing rerankers and the achievable upper bound. Further analysis reveals a critical distributional mismatch: while optimal predictions require evidence scattered across multiple languages, current systems systematically suppress such \`\`answer-critical'' documents, thereby limiting downstream generation performance. To bridge this gap, we propose \\textit{\\textbf{L}anguage-\\textbf{A}gnostic \\textbf{U}tility-driven \\textbf{R}eranker \\textbf{A}lignment (LAURA)}, which aligns multilingual evidence ranking with downstream generative utility. Experiments across diverse languages and generation models show that LAURA effectively mitigates language bias and consistently improves mRAG performance.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Downstream Generation and Ranking Alignment on MKQA (test)](<https://www.sota2.com/research/sota/downstream-generation-and-ranking-alignment-on-mkqa-test.md>): Downstream Generation and Ranking Alignment; MKQA (test); 3-gram Recall: 49.9.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.20199>)

[PDF](<https://arxiv.org/pdf/2604.20199.pdf>)