Jul 2026· Information Development· 0 citations· 45 references
TL;DR
The results show that English retrieval is already strong under BM25, whereas translated isiZulu queries benefit more clearly from multilingual semantic retrieval, especially at Top-10, and suggest that multilingual semantic retrieval can partly reduce translated-query barriers in open repository access.
Abstract
This article evaluates multilingual information access in three selected South African open scholarly repositories through a four-stage analytical heuristic: manual retrieval, keyword retrieval, multilingual semantic retrieval, and retrieval-augmented generation (RAG)-based question answering. The framework serves as an organizing device for comparing access mechanisms across conceptual, retrieval, and answer-generation layers. Using 4944 open scholarly records, 400 English queries, 400 machine-translated isiZulu queries, and 800 relevance judgments, the study compares BM25 keyword retrieval, multilingual-E5-large semantic retrieval, and a Qwen2.5-based RAG pipeline. The isiZulu component is interpreted as a translated-query benchmark rather than as evidence of authentic isiZulu user search behaviour. The results show that English retrieval is already strong under BM25, whereas translated isiZulu queries benefit more clearly from multilingual semantic retrieval, especially at Top-10. However, these gains remain bounded by the three-repository case-study corpus and do not remove cross-language access gaps. The RAG stage does not improve retrieval ranking beyond dense retrieval, but it changes the form of service by transforming retrieved evidence into natural-language answers. Answer evaluation indicates that larger generator scale improves faithfulness and evidence support, while completeness and isiZulu language quality remain limited. The findings suggest that multilingual semantic retrieval can partly reduce translated-query barriers in open repository access, but RAG-based library services require transparent evidence display, uncertainty signalling, and professional governance.
A novel benchmark framework for linguistic QA retrieval, empirical evidence supporting monolingual IR-specialised models, and insights into retrieval robustness under paraphrastic variation are included, enabling improved QA systems for specialised and low-resource environments.
Pedro Moura, Inês Gama, F. Batista et al.· 0 citations
Reasoning-intensive retrieval has become increasingly important for Retrieval-Augmented Generation (RAG), where systems must retrieve evidence that supports an answer rather than rely on surface matching alone. However, existing benchmarks mainly focus on English and leave Japanese retrieval underexplored. We introduce JaReIR, a Japanese benchmark built from Yahoo! Chiebukuro question-answer pairs and Japanese Wikipedia passages with crowdsourced relevance and answerability judgments. JaReIR contains 153 queries split into Easy queries, which have at least one Direct answer passage, and Hard queries, whose positive passages all require reasoning. We benchmark sparse and dense retrievers on JaReIR and analyze the effects of query expansion and reranking. Performance generally drops from Easy to Hard across retrieval settings, which shows that Japanese reasoning-intensive retrieval remains challenging. Passage-level analyses and retrieved-title dispersion statistics show that Hard queries yield more ambiguous top-ranked results. Document-level retrieval reduces, but does not eliminate, the Easy-Hard gap, suggesting that passage localization explains part of the difficulty. Evaluating a translated Japanese version of an existing English benchmark under the same passage-level setup further suggests that translation noise and linguistic differences contribute to lower performance.
This work proposes EnSI-RAG (Entity-Structure-Indexed Retrieval-Augmented Generation), a framework that constructs a query-independent, entity-centered index that separates evidence localization from answer synthesis while preserving traceable source evidence.
Xuanyu Meng, Jiashuo Sun, Jash Parekh et al.· 0 citations
Lexical retrieval (BM25) captures exact keyword matches and weights terms by corpus-wide significance, but it is blind to the semantic vocabulary gap: when a relevant document phrases an answer differently from the query, BM25 never retrieves it, and no amount of downstream reranking or fusion can recover a document that was never in the candidate set. We present Cross-Encoder Query Expansion (CE-QE), which reads the per-token relevance attributions of a cross-encoder applied to top semantic search results, selects the terms the cross-encoder treats as decisive, and appends them to the BM25 query. Unlike classical pseudo-relevance feedback, which reuses BM25's own (possibly wrong) top results, CE-QE seeds expansion from the semantic retriever's results, avoiding self-reinforcing query drift. Unlike recent generative query expansion (HyDE, Query2doc), which prompts a large language model to hallucinate text from its parametric knowledge, every CE-QE expansion term is copied verbatim from a retrieved passage, so it cannot introduce vocabulary the corpus does not contain, and its only added cost is attribution extraction on a cross-encoder a hybrid pipeline already runs for reranking. On seven BEIR datasets, CE-QE improves lexical recall substantially where query and answer vocabulary diverge (e.g., NQ Recall@100 from 0.32 to 0.47), and its score-fusion variant (SESF) beats cross-encoder score fusion by 2.5% on Recall@100 and beats SPLADEv2 and ColBERTv2 by 5.3% and 4.6% on nDCG@10, while leaving the underlying BM25 index completely unmodified.
Adam Kahirov, Umesh Deshpande, S. Sundararaman· 0 citations
DocuMind is a fully offline, privacy-preserving, multilingual Document Question Answering system built on the Retrieval-Augmented Generation (RAG) architecture, enabling true cross-lingual retrieval without any translation step.
S. S, Sripalreddy· International Scientific Jou...· 0 citations
This paper presents a comparative evaluation of cross-lingual information retrieval (CLIR) methods for retrieving English government information using Sinhala and Tamil queries. Two CLIR paradigms are investigated: Query Translation (QT), employing Google Translate, NLLB, and mBART50, and Cross-Lingual Embeddings (CLE), using LaBSE, multilingual E5, and BGE-M3, with monolingual English retrieval as the baseline. Experiments are conducted on a human-verified benchmark comprising 500 Sinhala, Tamil, and English question-answer pairs derived from 1,699 segmented contexts from Sri Lanka's Government Information Center (GIC). Retrieval performance is evaluated using Recall@k (k = 1, 3, 5, 10, 15). Monolingual retrieval performs poorly (Recall@15<10%), whereas all CLIR approaches substantially improve retrieval accuracy. Among them, BGE-M3 achieves the highest Recall@15, reaching 96.2% for Sinhala-English and 95.6% for Tamil-English, outperforming the best QT approach (Google Translate: 92.4% and 93.0%) while avoiding translation overhead. These results demonstrate that multilingual embedding models provide a more effective and scalable solution for cross-lingual retrieval-augmented generation (RAG) in low-resource government domains.
