Elamur Research⌕
☰
home / papers / 2026 / The Hundzula Retreat-Based Infrastructure Model for African Natural Language Processing

The Hundzula Retreat-Based Infrastructure Model for African Natural Language Processing

Paper · 2026 · Johannes Sibeko, Seani Rananga, Neo N. Putini et al.

Summary

African indigenous languages lack data, trained people, and lasting local structures for language research. Most work produces standalone corpora or lexicons and pays little attention to the conditions that keep such resources growing. The authors argue that the recurring research retreat is itself an infrastructure and should count as a language resource. Their case is the Hundzula Retreat for NLP and Linguistics, held annually in South Africa since 2022. Using a selectively applied design-science case study, they describe the fifth edition, a four-day programme. It moved from linguistic complexity, such as polysemy in cross-border languages, to corpus and lexicon building, ethics and data governance, and applied systems including speech recognition, LLM deployment, and chatbots. The retreat mixes training, human-in-the-loop methods, mentoring, and incubation of projects that are still in progress. Its governance stresses credit attribution and shared ownership. The main contribution is conceptual: a framework other low-resourced language communities can adapt. The authors admit its transferability has not been tested empirically.

Key points

  • "Hundzula" comes from Xitsonga and means transformation. The retreat has run each February since 2022, hosted by the University of Pretoria (2022–2023), Nelson Mandela University (2024), University of Johannesburg (2025), and North-West University (2026).
  • The paper analyses the fifth edition, a four-day event covering the resource lifecycle from conceptualisation and annotation to modelling and application.
  • Selection deliberately favours ongoing projects. Examples include work on non-standard and contact varieties such as Sepitori and spoken Khelobedu, plus "living dictionaries" treated as evolving datasets.
  • Ethics, privacy, and community-sensitive data governance are discussed alongside technical work. The paper references African-oriented licensing efforts such as the NOODL licence.
  • Limitations: the framework is argued conceptually, not validated across settings. Future work should track how retreat-incubated projects become formal NLP resources.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓University of PretoriaPretoria · South Africa9 works · 755 citations🎓University of KwaZulu-NatalDurban · South Africa3 works · 15 citations

Also on the byline: Nelson Mandela University.

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 worksLiterature and Literary Theory · 541 works

Cite

Cite as Johannes Sibeko, Seani Rananga, Neo N. Putini and Dan Masethe. 2026. The Hundzula Retreat-Based Infrastructure Model for African Natural Language Processing. Proceedings of Resources for African Indigenous Languages (RAIL) 2026 @ LREC 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.63317/54dm3hjmmfdr

@inproceedings{sibeko2026hundzula,
  title = {{The Hundzula Retreat-Based Infrastructure Model for African Natural Language Processing}},
  author = {Johannes Sibeko, Seani Rananga, Neo N. Putini and Dan Masethe},
  booktitle = {Proceedings of Resources for African Indigenous Languages (RAIL) 2026 @ LREC 2026},
  year = {2026},
  doi = {10.63317/54dm3hjmmfdr},
  url = {https://doi.org/10.63317/54dm3hjmmfdr}
}

References

18 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Ifeoluwanimi Adebara. 2024. Towards Afrocentric natural language processing. Ph.D. thesis, University of British Columbia, Vancouver, BC Canada. find
  2. Emmanuel Kigen Chesire and Andrew Kipkebut. 2024. Current state, challenges and opportunities for natural language processing research and development in africa: A systematic review. In 5th Workshop on African Natural Language Processing. find
  3. Jeanne Elizabeth Daniel. 2020. Applications of natural language processing for low-resource languages in the healthcare domain. Ph.D. thesis, Stellenbosch: Stellenbosch University, South Africa. find
  4. Alladi Deekshith. 2024. Advances in natural language processing: A survey of techniques. International Journal of Innovations in Engineering Research and Technology, 8:74–83. find
  5. Anjalie Field, Shrimai Prabhumoye, Maarten Sap, Zhijing Jin, Jieyu Zhao, and Chris Brockett. 2021. Proceedings of the 1st workshop on nlp for positive impact. In Proceedings of the 1st Workshop on NLP for Positive Impact, Online. Association for Computational Linguistics. find
  6. Diksha Khurana, Aditya Koli, Kiran Khatter, and Sukhdev Singh. 2023. Natural language processing: state of the art, current trends and challenges. Multimedia tools and applications, 82(3):3713–3744. find
  7. Rooweither Mabuya, Don Mthobela, Mmasibidi Setaka, and Menno van Zaanen. 2023. Proceedings of the fourth workshop on resources for african indigenous languages (rail 2023). In Proceedings of the Fourth workshop on Resources for African Indigenous Languages (RAIL 2023). find
  8. Matthew S Mayernik, David L Hart, Keith E Maull, and Nicholas M Weber. 2017. Assessing and tracing the outcomes and impact of research infrastructures. Journal of the Association for Information Science and Technology, 68(6):1341– 1359. find
  9. Derguene Mbaye, Tatiana DP Mbengue, Madoune R Seye, Moussa Diallo, Mamadou L Ndiaye, Dimitri S Adjanohoun, Cheikh S Wade, Djiby Sow, Jean-Claude B Munyaka, and Jerome Chenal. 2025. Opportunities and challenges of natural language processing for low-resource senegalese languages in social science res… — [arXiv:2601.09716](https://arxiv.org/abs/2601.09716) find
  10. Shamsuddeen Hassan Muhammad, Ibrahim Sa’id Ahmad, Idris Abdulmumin, Falalu Ibrahim Lawan, Sukairaj Hafiz Imam, Yusuf Aliyu, Sani Abdullahi Sani, Ali Usman Umar, Tajuddeen Gwadabe, Kenneth Church, et al. 2025. Hausanlp: Current status, challenges and future directions for hausa natural language proce… find
  11. Nadia Nahar, Shurui Zhou, Grace A. Lewis, and Christian Kästner. 2021. More engineering, no silos: Rethinking processes and interfaces in collaboration between interdisciplinary teams for machine learning projects. ArXiv, abs/2110.10234. — [arXiv:2110.10234](https://arxiv.org/abs/2110.10234) find
  12. Tolúlo.pé. Ògúnré.mí, Wilhelmina Onyothi Nekoto, and Saron Samuel. 2023. Decolonizing nlp for “low-resource languages”: Applying abebe birhane’s relational ethics. GRACE: Global Review of AI Community Ethics, 1(1):1–13. find
  13. Chijioke Okorie. 2023. Copyright, data mining and developing models for south african natural language processing. Joint PIJIP/TLS Research Paper Series, 117:1–28. find
  14. Chijioke Okorie. 2025. It’s the NOODL license– awesome and amazingly geeky! Available at SSRN https://ssrn.com/abstract=5339254. ↗
  15. Chijioke Okorie and Melissa Omino. 2025. Addressing inequitable openness in licences for sharing african data and datasets through the nwulite obodo open data licence. Law, Tech. & Hum., 7:94. find
  16. Mmasibidi Setaka and Benito Trollip. 2022. Resource repositories and linking resources: An exploratory study. Journal of the Digital Humanities Association of Southern Africa (DHASA), 4(02). find
  17. Kathleen Siminyu, Jade Abbott, Kola Tubø.sun, Aremu Anuoluwapo, Blessing K Sibanda, Kofi Yeboah, David Adelani, Masabata Mokgesi-Selinga, Frederick R Apina, Angela Thandizwe Mthembu, et al. 2023. Consultative engagement of stakeholders toward a roadmap for african language technologies. Patterns, 4(… find
  18. Tianyang Zhong, Zhenyuan Yang, Zhengliang Liu, Ruidong Zhang, Weihang You, Yiheng Liu, Haiyang Sun, Yi Pan, Yiwei Li, Yifan Zhou, et al. 2024. Opportunities and challenges of large language models for low-resource languages in humanities research. arXiv preprint arXiv:2412.04497. — [arXiv:2412.04497](https://arxiv.org/abs/2412.04497) find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/sibeko-2026-hundzula-retreat-based-infrastructure-model/