Multilingual Retrieval-Augmented Generation for Knowledge-Intensive Question Answering Task
Open accessAbstract
Retrieval-augmented generation (RAG) has become a cornerstone of contemporary NLP, enhancing large language models (LLMs) by allowing them to access richer factual contexts through in-context retrieval.While effective in monolingual settings, especially in English, its use in multilingual tasks remains unexplored.This paper investigates the effectiveness of RAG across multiple languages by proposing novel approaches for multilingual opendomain question-answering.We evaluate the performance of various multilingual RAG strategies, including question-translation (tRAG), which translates questions into English before retrieval, and Multilingual RAG (MultiRAG), where retrieval occurs directly across multiple languages.Our findings reveal that tRAG, while useful, suffers from limited coverage.In contrast, MultiRAG improves efficiency by enabling multilingual retrieval but introduces inconsistencies due to crosslingual variations in the retrieved content.To address these issues, we propose Crosslingual RAG (CrossRAG), a method that translates retrieved documents into a common language (e.g., English) before generating the response.Our experiments show that CrossRAG significantly enhances performance on knowledgeintensive tasks, benefiting both high-resource and low-resource languages.Question Il principio di "uniformità della natura" è un framework teorico sviluppato da Aristotele.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Translating queries and retrieved contexts into a high-resource language can substantially enhance generation quality in RAG pipelines.
“This approach builds upon recent empirical findings demonstrating that translating queries and retrieved contexts into a high-resource language can substantially enhance generation quality in RAG pipelines (Ranaldi et al., 2025).”
From Improving Amharic Information Retrieval with Translative and Multi-Agent Debate Retrieval Augmented Generation (2026) · cites this paper as Ranaldi et al., 2025
Section 4.3 reports that translating retrieved information into English improves CrossRAG performance. But the paper evaluates query translation separately as tRAG and reports no conspicuous improvements from it; the Introduction also warns that incorrect query translations can lead to wrong retrieval and misleading generations. It does not demonstrate that translating both queries and retrieved contexts enhances generation quality. (LLM, conf=0.88)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
No references could be resolved yet.
Research works printed but not linked: 20
- Akari Asai, Jungo Kasai, Jonathan Clark, Kenton Lee, Eunsol Choi, and Hannaneh Hajishirzi. 2021. XOR QA: Cross-lingual open-retrieval question answering. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computa- tional Linguistics: Human Language Technologies, pages 547-564, Online. Association for Computa- tional Linguistics. · DOI
- Akari Asai, Sewon Min, Zexuan Zhong, and Danqi Chen. 2023. Retrieval-based language models and applications. In Proceedings of the 61st Annual Meeting of the Association for Computational Lin- guistics (Volume 6: Tutorial Abstracts), pages 41-46, Toronto, Canada. Association for Computational Lin- guistics. · DOI · Paper page
- Laurie Burchell, Alexandra Birch, Nikolay Bogoychev, and Kenneth Heafield. 2023. An open dataset and model for language identification. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 865-879, Toronto, Canada. Association for Computational Linguistics. · DOI
- Jonathan H. Clark, Eunsol Choi, Michael Collins, Dan Garrette, Tom Kwiatkowski, Vitaly Nikolaev, and Jennimaria Palomaki. 2020. TyDi QA: A benchmark for information-seeking question answering in ty- pologically diverse languages. Transactions of the Association for Computational Linguistics, 8:454- 470. · DOI
- Luyu Gao, Zhuyun Dai, Panupong Pasupat, Anthony Chen, Arun Tejasvi Chaganty, Yicheng Fan, Vincent Zhao, Ni Lao, Hongrae Lee, Da-Cheng Juan, and Kelvin Guu. 2023. RARR: Researching and revising what language models say, using language models. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 16477-16508, Toronto, Canada. Association for Computational Linguistics. · DOI · Paper page
- Tom Kwiatkowski, Jennimaria Palomaki, Olivia Red- field, Michael Collins, Ankur Parikh, Chris Al- berti, Danielle Epstein, Illia Polosukhin, Jacob De- vlin, Kenton Lee, Kristina Toutanova, Llion Jones, Matthew Kelcey, Ming-Wei Chang, Andrew M. Dai, Jakob Uszkoreit, Quoc Le, and Slav Petrov. 2019. Natural questions: A benchmark for question an- swering research. Transactions of the Association for Computational Linguistics, 7:452-466. · DOI · Paper page
- Patrick Lewis, Barlas Oguz, Ruty Rinott, Sebastian Riedel, and Holger Schwenk. 2020a. MLQA: Eval- uating cross-lingual extractive question answering. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7315-7330, Online. Association for Computational Linguistics. · DOI
- Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Hein- rich Küttler, Mike Lewis, Wen-tau Yih, Tim Rock- täschel, et al. 2020b. Retrieval-augmented genera- tion for knowledge-intensive nlp tasks. Advances in Neural Information Processing Systems, 33:9459- 9474.
- Alex Mallen, Akari Asai, Victor Zhong, Rajarshi Das, Daniel Khashabi, and Hannaneh Hajishirzi. 2023. When not to trust language models: Investigating effectiveness of parametric and non-parametric mem- ories. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Vol- ume 1: Long Papers), pages 9802-9822, Toronto, Canada. Association for Computational Linguistics. · DOI · Paper page
- Giulia Pucci and Leonardo Ranaldi. 2025. Advancing oversight reasoning across languages for audit syco- phantic behaviour via X-agent. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 12949-12965, Suzhou, China. Association for Computational Linguistics. · DOI
- Pranav Rajpurkar, Jian Zhang, Konstantin Lopyrev, and Percy Liang. 2016. SQuAD: 100,000+ questions for machine comprehension of text. In Proceedings of the 2016 Conference on Empirical Methods in Natu- ral Language Processing, pages 2383-2392, Austin, Texas. Association for Computational Linguistics. · DOI · Paper page
- Federico Ranaldi, Andrea Zugarini, Leonardo Ranaldi, and Fabio Massimo Zanzotto. 2025a. Protoknowl- edge shapes behaviour of llms in downstream tasks: Memorization and generalization with knowledge graphs. · DOI
- Leonardo Ranaldi and Andre Freitas. 2024. Align- ing large and small language models via chain-of- thought reasoning. In Proceedings of the 18th Con- ference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Pa- pers), pages 1812-1827, St. Julian's, Malta. Associ- ation for Computational Linguistics. · DOI
- Leonardo Ranaldi, Barry Haddow, and Alexandra Birch. 2025b. When natural language is not enough: The limits of in-context learning demonstrations in mul- tilingual reasoning. In Findings of the Association for Computational Linguistics: NAACL 2025, pages 7369-7396, Albuquerque, New Mexico. Association for Computational Linguistics. · DOI
- Leonardo Ranaldi and Giulia Pucci. 2025. Multilin- gual reasoning via self-training. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Lin- guistics: Human Language Technologies (Volume 1: Long Papers), pages 11566-11582, Albuquerque, New Mexico. Association for Computational Lin- guistics. · DOI
- Leonardo Ranaldi, Giulia Pucci, Barry Haddow, and Alexandra Birch. 2024. Empowering multi-step rea- soning across languages via program-aided language models. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 12171-12187, Miami, Florida, USA. Associa- tion for Computational Linguistics. · DOI
- Leonardo Ranaldi, Marco Valentino, and Andre Fre- itas. 2025d. Eliciting critical reasoning in retrieval- augmented generation via contrastive explanations. In Proceedings of the 2025 Conference of the Na- tions of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 11168- 11183, Albuquerque, New Mexico. Association for Computational Linguistics.
- Timo Schick, Jane Dwivedi-Yu, Roberto Dessì, Roberta Raileanu, Maria Lomeli, Luke Zettlemoyer, Nicola Cancedda, and Thomas Scialom. 2023. Toolformer: Language models can teach themselves to use tools. arXiv preprint arXiv:2302.04761. · arXiv
- Nikhil Sharma, Kenton Murray, and Ziang Xiao. 2024. Faux polyglot: A study on information disparity in multilingual large language models. · DOI · Paper page
- Jörg Tiedemann. 2012. Parallel data, tools and inter- faces in OPUS. In Proceedings of the Eighth In- ternational Conference on Language Resources and Evaluation (LREC'12), pages 2214-2218, Istanbul, Turkey. European Language Resources Association (ELRA).
Cited by
Sorted by citations
Jotie, A. A., & Mitra, P. (2026) · Proceedings of the Language Resources and Evaluation Conference · doi:10.63317/3hbm9z4okcrx
- Authors
- Ranaldi, L.
- Affiliations
- University of Edinburgh (GB)
- Corresponding author
- Ranaldi, Leonardo
- ORCID
- Ranaldi, Leonardo
- Venue
- Findings of the Association for Computational Linguistics: EACL 2026
- Published
- 2026
- Pages
- 697-716
- OpenAlex
- W4414957389
- Cited by
- 22
- References
- 0
- Citations by year
- 2026: 17 · 2025: 5
- FWCI
- 70.79
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Findings of the Association for Computational Linguistics: EACL (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion, cc-by)
- Edinburgh Research Explorer (University of Edinburgh) (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Funders
- Government of the United Kingdom
- UK Research and Innovation (10039436)
- European Commission (101070631)
- HORIZON EUROPE Framework Programme (101070631)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Information Systems › Information Retrieval and Search Behavior
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Keywords
- retrieval-augmented generation, multilingual question answering, open-domain question answering, cross-lingual information retrieval, question translation, document translation, knowledge-intensive question answering, low-resource languages, large language models
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
