Linguistic Nepotism: Trading-off Quality for Language Preference in Multilingual RAG
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914071183360000 |
|---|---|
| author | Ki, Dayeon Carpuat, Marine McNamee, Paul Khashabi, Daniel Yang, Eugene Lawrie, Dawn Duh, Kevin |
| author_facet | Ki, Dayeon Carpuat, Marine McNamee, Paul Khashabi, Daniel Yang, Eugene Lawrie, Dawn Duh, Kevin |
| contents | Multilingual Retrieval-Augmented Generation (mRAG) systems enable language models to answer knowledge-intensive queries with citation-supported responses across languages. While such systems have been proposed, an open questions is whether the mixture of different document languages impacts generation and citation in unintended ways. To investigate, we introduce a controlled methodology using model internals to measure language preference while holding other factors such as document relevance constant. Across eight languages and six open-weight models, we find that models preferentially cite English sources when queries are in English, with this bias amplified for lower-resource languages and for documents positioned mid-context. Crucially, we find that models sometimes trade-off document relevance for language preference, indicating that citation choices are not always driven by informativeness alone. Our findings shed light on how language models leverage multilingual context and influence citation behavior. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_13930 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Linguistic Nepotism: Trading-off Quality for Language Preference in Multilingual RAG Ki, Dayeon Carpuat, Marine McNamee, Paul Khashabi, Daniel Yang, Eugene Lawrie, Dawn Duh, Kevin Computation and Language Multilingual Retrieval-Augmented Generation (mRAG) systems enable language models to answer knowledge-intensive queries with citation-supported responses across languages. While such systems have been proposed, an open questions is whether the mixture of different document languages impacts generation and citation in unintended ways. To investigate, we introduce a controlled methodology using model internals to measure language preference while holding other factors such as document relevance constant. Across eight languages and six open-weight models, we find that models preferentially cite English sources when queries are in English, with this bias amplified for lower-resource languages and for documents positioned mid-context. Crucially, we find that models sometimes trade-off document relevance for language preference, indicating that citation choices are not always driven by informativeness alone. Our findings shed light on how language models leverage multilingual context and influence citation behavior. |
| title | Linguistic Nepotism: Trading-off Quality for Language Preference in Multilingual RAG |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2509.13930 |