Lugha-Llama: Adapting Large Language Models for African Languages
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908308684668928 |
|---|---|
| author | Buzaaba, Happy Wettig, Alexander Adelani, David Ifeoluwa Fellbaum, Christiane |
| author_facet | Buzaaba, Happy Wettig, Alexander Adelani, David Ifeoluwa Fellbaum, Christiane |
| contents | Large language models (LLMs) have achieved impressive results in a wide range of natural language applications. However, they often struggle to recognize low-resource languages, in particular African languages, which are not well represented in large training corpora. In this paper, we consider how to adapt LLMs to low-resource African languages. We find that combining curated data from African languages with high-quality English educational texts results in a training mix that substantially improves the model's performance on these languages. On the challenging IrokoBench dataset, our models consistently achieve the best performance amongst similarly sized baselines, particularly on knowledge-intensive multiple-choice questions (AfriMMLU). Additionally, on the cross-lingual question answering benchmark AfriQA, our models outperform the base model by over 10%. To better understand the role of English data during training, we translate a subset of 200M tokens into Swahili language and perform an analysis which reveals that the content of these data is primarily responsible for the strong performance. We release our models and data to encourage future research on African languages. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_06536 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Lugha-Llama: Adapting Large Language Models for African Languages Buzaaba, Happy Wettig, Alexander Adelani, David Ifeoluwa Fellbaum, Christiane Computation and Language Artificial Intelligence Machine Learning Large language models (LLMs) have achieved impressive results in a wide range of natural language applications. However, they often struggle to recognize low-resource languages, in particular African languages, which are not well represented in large training corpora. In this paper, we consider how to adapt LLMs to low-resource African languages. We find that combining curated data from African languages with high-quality English educational texts results in a training mix that substantially improves the model's performance on these languages. On the challenging IrokoBench dataset, our models consistently achieve the best performance amongst similarly sized baselines, particularly on knowledge-intensive multiple-choice questions (AfriMMLU). Additionally, on the cross-lingual question answering benchmark AfriQA, our models outperform the base model by over 10%. To better understand the role of English data during training, we translate a subset of 200M tokens into Swahili language and perform an analysis which reveals that the content of these data is primarily responsible for the strong performance. We release our models and data to encourage future research on African languages. |
| title | Lugha-Llama: Adapting Large Language Models for African Languages |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2504.06536 |