AfriMTEB and AfriE5: Benchmarking and Adapting Text Embedding Models for African Languages
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915838339055616 |
|---|---|
| author | Uemura, Kosei Zhang, Miaoran Adelani, David Ifeoluwa |
| author_facet | Uemura, Kosei Zhang, Miaoran Adelani, David Ifeoluwa |
| contents | Text embeddings are an essential building component of several NLP tasks such as retrieval-augmented generation which is crucial for preventing hallucinations in LLMs. Despite the recent release of massively multilingual MTEB (MMTEB), African languages remain underrepresented, with existing tasks often repurposed from translation benchmarks such as FLORES clustering or SIB-200. In this paper, we introduce AfriMTEB -- a regional expansion of MMTEB covering 59 languages, 14 tasks, and 38 datasets, including six newly added datasets. Unlike many MMTEB datasets that include fewer than five languages, the new additions span 14 to 56 African languages and introduce entirely new tasks, such as hate speech detection, intent detection, and emotion classification, which were not previously covered. Complementing this, we present AfriE5, an adaptation of the instruction-tuned mE5 model to African languages through cross-lingual contrastive distillation. Our evaluation shows that AfriE5 achieves state-of-the-art performance, outperforming strong baselines such as Gemini-Embeddings and mE5. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_23896 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | AfriMTEB and AfriE5: Benchmarking and Adapting Text Embedding Models for African Languages Uemura, Kosei Zhang, Miaoran Adelani, David Ifeoluwa Computation and Language Text embeddings are an essential building component of several NLP tasks such as retrieval-augmented generation which is crucial for preventing hallucinations in LLMs. Despite the recent release of massively multilingual MTEB (MMTEB), African languages remain underrepresented, with existing tasks often repurposed from translation benchmarks such as FLORES clustering or SIB-200. In this paper, we introduce AfriMTEB -- a regional expansion of MMTEB covering 59 languages, 14 tasks, and 38 datasets, including six newly added datasets. Unlike many MMTEB datasets that include fewer than five languages, the new additions span 14 to 56 African languages and introduce entirely new tasks, such as hate speech detection, intent detection, and emotion classification, which were not previously covered. Complementing this, we present AfriE5, an adaptation of the instruction-tuned mE5 model to African languages through cross-lingual contrastive distillation. Our evaluation shows that AfriE5 achieves state-of-the-art performance, outperforming strong baselines such as Gemini-Embeddings and mE5. |
| title | AfriMTEB and AfriE5: Benchmarking and Adapting Text Embedding Models for African Languages |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2510.23896 |