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

• URL: http://arxiv.org/abs/2604.20199v1
• Date: Wed, 22 Apr 2026 05:33:06 GMT
• Title: All Languages Matter: Understanding and Mitigating Language Bias in Multilingual RAG
• Abstract summary: We show that current Multilingual Retrieval-Augmented Generation (mRAG) systems suffer from a language bias during reranking.<n>We propose textittextbfLanguage-textbfAgnostic textbfUtility-driven textbfReranker textbfAlignment (LAURA), which aligns multilingual evidence ranking with downstream generative utility.
• Score: 87.74635133954975
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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.

Related papers

• Inference-Time Steering for Cross-Lingual Factual Consistency in LLMs [51.56484100374058]
  Large Language Models (LLMs) demonstrate remarkable multilingual fluency, but their internal knowledge representations remain disproportionately biased toward high-resource languages.<n>We investigate whether these biases can be mitigated at inference time, forcing an English-prompted model to answer as if it were queried in target languages.<n>Experiments on Gemma 3 12B Instruct reveal persona prompting to be the strongest overall intervention, balancing efficacy, safety, and out-of-domain generalization.
  arXiv  Detail & Related papers  (2026-07-21T16:15:05Z)
• Enhancing Multilingual RAG Systems with Debiased Language Preference-Guided Query Fusion [12.496203909541721]
  MRAG systems often exhibit a perceived preference for high-resource languages, particularly English.<n>We find that such measurements are significantly distorted by structural priors inherent in evaluation benchmarks.<n>We propose DeLP (Debiased Language Preference), a calibrated metric designed to explicitly factor out these structural confounds.
  arXiv  Detail & Related papers  (2026-01-06T12:01:56Z)
• Parallel Universes, Parallel Languages: A Comprehensive Study on LLM-based Multilingual Counterfactual Example Generation [49.2073409243885]
  Large language models (LLMs) excel at generating English counterfactuals and demonstrate multilingual proficiency.<n>We conduct automatic evaluations on both directly generated counterfactuals in the target languages and those derived via English translation across six languages.<n>We identify and categorize four main types of errors that consistently appear in the generated counterfactuals across languages.
  arXiv  Detail & Related papers  (2026-01-01T08:53:49Z)
• Language Drift in Multilingual Retrieval-Augmented Generation: Characterization and Decoding-Time Mitigation [11.110312833458421]
  We study output language drift in multilingual RAG across multiple datasets, languages, and LLM backbones.<n>Our experiments reveal that the drift results from decoder-level collapse, where dominant token distributions and high-frequency English patterns dominate the intended generation language.<n>We propose Soft Constrained Decoding (SCD), a lightweight, training-free decoding strategy that gently steers generation toward the target language.
  arXiv  Detail & Related papers  (2025-11-13T05:36:31Z)
• Investigating Language and Retrieval Bias in Multilingual Previously Fact-Checked Claim Detection [4.6738956348193]
  Large Language Models (LLMs) offer powerful capabilities for cross-lingual fact-checking.<n>LLMs often exhibit language bias, performing disproportionately better on high-resource languages such as English.<n>We present and inspect a novel concept - retrieval bias, when information retrieval systems tend to favor certain information over others.
  arXiv  Detail & Related papers  (2025-09-29T17:50:32Z)
• Linguistic Nepotism: Trading-off Quality for Language Preference in Multilingual RAG [55.258582772528506]
  We investigate whether the mixture of different document languages impacts generation and citation in unintended ways.<n>Across eight languages and six open-weight models, we find that models preferentially cite English sources when queries are in English.<n>We find that models sometimes trade-off document relevance for language preference, indicating that citation choices are not always driven by informativeness alone.
  arXiv  Detail & Related papers  (2025-09-17T12:58:18Z)
• Multilingual Retrieval-Augmented Generation for Knowledge-Intensive Task [89.45111250272559]
  Retrieval-augmented generation (RAG) has become a cornerstone of contemporary NLP.<n>This paper investigates the effectiveness of RAG across multiple languages by proposing novel approaches for multilingual open-domain question-answering.
  arXiv  Detail & Related papers  (2025-04-04T17:35:43Z)
• Investigating Language Preference of Multilingual RAG Systems [4.438698005789677]
  MRAG systems struggle with retrieving relevant information due to linguistic variations between queries and documents.<n>We propose Dual Knowledge Multilingual RAG (DKM-RAG), a framework that fuses translated multilingual passages with complementary model knowledge.<n> Empirical results demonstrate that DKM-RAG mitigates language preference in generation and enhances performance across diverse linguistic settings.
  arXiv  Detail & Related papers  (2025-02-16T15:54:05Z)
• The Power of Question Translation Training in Multilingual Reasoning: Broadened Scope and Deepened Insights [108.40766216456413]
  We propose a question alignment framework to bridge the gap between large language models' English and non-English performance. Experiment results show it can boost multilingual performance across diverse reasoning scenarios, model families, and sizes. We analyze representation space, generated response and data scales, and reveal how question translation training strengthens language alignment within LLMs.
  arXiv  Detail & Related papers  (2024-05-02T14:49:50Z)
• On Cross-Lingual Retrieval with Multilingual Text Encoders [51.60862829942932]
  We study the suitability of state-of-the-art multilingual encoders for cross-lingual document and sentence retrieval tasks. We benchmark their performance in unsupervised ad-hoc sentence- and document-level CLIR experiments. We evaluate multilingual encoders fine-tuned in a supervised fashion (i.e., we learn to rank) on English relevance data in a series of zero-shot language and domain transfer CLIR experiments.
  arXiv  Detail & Related papers  (2021-12-21T08:10:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.