Benchmarking Retrieval-Augmented Generation for Scientific Knowledge QA in European Portuguese

Jose Matos, Catarina Silva, Hugo Goncalo Oliveira


Abstract
Retrieval-Augmented Generation (RAG) enables grounding of model outputs in external evidence, but its impact on European Portuguese (pt-PT) scientific question answering (QA) remains unclear. We present a controlled evaluation of RAG on pt-PT knowledge QA across different scientific domains using the Portuguese test split of the Global MMLU Lite dataset. As external evidence, we use a Portuguese scientific literature knowledge base containing over 32,000 documents converted to Markdown. We benchmark five instruction-tuned small language models (4-12B) and compare closed-book baselines against 16 RAG configurations that vary by: (i) dense retriever specialization (multilingual vs. Portuguese-specific), (ii) reranking (on/off), and (iii) number of retrieved chunks (k ∈ 1, 3, 5, 10). Results suggest that RAG gains are model-dependent. Some models improve consistently, others are highly sensitive to retrieval choices, and some degrade under retrieval noise, especially at larger values of k. Findings highlight the importance of model-specific retrieval tuning and ensuring that the retriever and reranker languages and domains align when deploying RAG systems for Portuguese natural scientific language processing.
Anthology ID:
2026.nslp-1.3
Volume:
Proceedings of Natural Scientific Language Processing (NSLP) @ LREC 2026
Month:
May
Year:
2026
Address:
Palma, Mallorca (Spain)
Editors:
Georg Rehm, Stefan Dietze, Danilo Dessi, Diana Maynard, Sonja Schimmler
Venues:
NSLP | WS
SIG:
Publisher:
ELRA Language Resources Association (ELRA)
Note:
Pages:
25–31
Language:
External URL:
https://lrec.elra.info/lrec2026-ws-nslp-03
DOI:
10.63317/3muergicuxwk
Bibkey:
Cite (ACL):
Jose Matos, Catarina Silva, and Hugo Goncalo Oliveira. 2026. Benchmarking Retrieval-Augmented Generation for Scientific Knowledge QA in European Portuguese. In Proceedings of Natural Scientific Language Processing (NSLP) @ LREC 2026, pages 25–31, Palma, Mallorca (Spain). ELRA Language Resources Association (ELRA).
Cite (Informal):
Benchmarking Retrieval-Augmented Generation for Scientific Knowledge QA in European Portuguese (Matos et al., NSLP 2026)
Copy Citation: