mRAKL: Multilingual Retrieval-Augmented Knowledge Graph Construction for Low-Resourced Languages

2025-07-21

AI Summary

This work addresses multilingual knowledge graph completion (mKGC) for low-resource languages—specifically Tigrinya and Amharic—by reformulating mKGC as a cross-lingual question answering (QA) task and proposing a retrieval-augmented generation (RAG)-based solution. Methodologically, a BM25 retriever fetches relevant context from English/Arabic knowledge sources, which then conditions a multilingual QA-style generative model to predict missing entities and relations, enabling cross-lingual knowledge transfer. Key contributions include: (i) the first application of the RAG paradigm to low-resource mKGC; (ii) empirical validation that an ideal retriever yields substantial accuracy gains; and (iii) performance improvement without any target-language annotated data. Experiments demonstrate that RAG achieves absolute accuracy improvements of +4.92 and +8.79 percentage points over context-free baselines on Tigrinya and Amharic, respectively—marking significant progress in low-resource knowledge graph construction.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Knowledge Representation and Reasoning: Knowledge Representation Languages
• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Knowledge Graphs represent real-world entities and the relationships between them. Multilingual Knowledge Graph Construction (mKGC) refers to the task of automatically constructing or predicting missing entities and links for knowledge graphs in a multilingual setting. In this work, we reformulate the mKGC task as a Question Answering (QA) task and introduce mRAKL: a Retrieval-Augmented Generation (RAG) based system to perform mKGC. We achieve this by using the head entity and linking relation in a question, and having our model predict the tail entity as an answer. Our experiments focus primarily on two low-resourced languages: Tigrinya and Amharic. We experiment with using higher-resourced languages Arabic and English for cross-lingual transfer. With a BM25 retriever, we find that the RAG-based approach improves performance over a no-context setting. Further, our ablation studies show that with an idealized retrieval system, mRAKL improves accuracy by 4.92 and 8.79 percentage points for Tigrinya and Amharic, respectively.

Problem

• Constructs multilingual knowledge graphs for low-resourced languages.
• Reformulates knowledge graph construction as question answering.
• Improves accuracy using retrieval-augmented generation for Tigrinya and Amharic.

Innovation

• Reformulates mKGC as QA task
• Uses Retrieval-Augmented Generation (RAG)
• Focuses on low-resourced languages