Bridging the Gap: Cross-Lingual Transfer Learning to Address Data Scarcity in Igbo Dialects for Machine Translation

Transfer learning is a highly effective solution for data-scarce indigenous languages, overcoming the severe data limitations they face. By leveraging pre-trained models from high-resource languages, algorithms adapt to target indigenous languages using minimal data. This approach achieves high performance in machine translation and semantic understanding without needing massive, expensive, original datasets.   To bridge the digital language gap and ensure accurate, localised dialect representation in AI, researchers should adopt a community-centric participatory framework. This includes using native speakers to crowdsource localised texts and leveraging advanced foundational models to scale these efforts.

Abdulyekeen, R. & Abdullahi, B. A. (2026). Cross-Lingual Sentiment Analysis with Natural Language Processing: Insights from Selected Nigeria Languages (Yoruba, Hausa, Igbo, and Nigerian Pidgin). Journal of Multidisciplinary Science: MIKAILALSYS, 4(1), 60-86. DOI: 10.58578/mikailalsys.v4i1.8596 Adebara, I., Elmadany, A., Abdul-Mageed, M., & Alcoba, A. (2023). SERENGETI: Massively Multilingual Language Models for Africa. Findings of the Association for Computational Linguistics: ACL 2023, 1498-1537. DOI:10.18653/v1/2023.findings-acl.97 Agbeyangi, A. O., Eludiora, S. I. & Adenekan, A. O. (2015). English to Yorúbà Machine Translation System Using Rule-Based Approach. Journal of Multidisciplinary Engineering Science and Technology, 2(8), 2275-2280. https://www.jmest.org/wp-content/uploads/JMESTN42351018.pdf Akwanya, A. N. (2025). Nigerian English Writing and Translation: The Fate of the Vernacular Literature and Culture. Journal of Creative Research in English Literature and Culture (JCRELC), 1(3), 16-25. Available at: https://jcrelc.com/assets/articles_menuscripts/file/JCRELC-NOV-2025-3.pdf Alabi, J. O., Adelani, D. I., Mosbach, M., & Klakow, D. (2022, Oct 18). Adapting Pre-trained Language Models to African Languages via Multilingual Adaptive Fine-Tuning. https://arxiv.org/abs/2204.06487 Artiba (2025, Jun 4). Synthetic Data in AI: Benefits, Use Cases & Impact. https://www.artiba.org/blog/synthetic-data-in-ai-benefits-use-cases-and-impact Asprovska, M. and Hunter, N. (2024). The Tokenization Problem: Understanding Generative AI's Computational Language Bias. Ubiquity Proceedings, Article 1. DOI: 10.5334/uproc.123 Ayogu, I. I., Adetunmbi, A. O., & Ojokoh, B. O. (2018). Developing Statistical Machine Translation System for English and Nigerian Languages. Asian Journal of Research in Computer Science 1(4), 1-8. Belay, T. D., Azime, I. A., Ahmad, I. S., Adelani, D. I., Abdulmumin, I., Ayele, A. A., Muhammad, S. H., & Yimam, S. M. (2025). AfroXLMR-Social: Adapting Pre-trained Language Models for African Languages Social Media Text. In Findings of the Association for Computational Linguistics: EMNLP 2025, 15570–15587, Suzhou, China. Bilali, H. (2025, Sep 22). Nigeria Launches N-ATLAS, an AI Language Model in Yoruba, Igbo, Hausa, and Nigerian English. https://www.wearetech.africa/en/fils-uk/news/tech/nigeria-launches-n-atlas-an-ai-language-model-in-yoruba-igbo-hausa-and-nigerian-english Boujkian, S. (2025). Improving Low-Resource Machine Translation via Cross-Linguistic Transfer from Typologically Similar High-Resource Languages. https://arxiv.org/abs/2501.00045