Where Are We? Evaluating LLM Performance on African Languages
Ife Adebara, Hawau Olamide Toyin, Nahom Tesfu Ghebremichael, AbdelRahim A. Elmadany, Muhammad Abdul-Mageed
Abstract
Language policies for education across Africa. (b) Official national language policies across Africa. (c) Official regional language policies across Africa. None means no additional policy is available on a regional level. (d) We collect SAHARA to empirically evaluate LLM performance on African languages, allowing us to demonstrate with evidence how current language policies directly impact progress in the field. Figure 1: Maps of Africa showing the languages covered in this work and the language policies across the continent. The term Indigenous language is broadly defined here as one that is 'native' to the area (Walsh, 2005). The term Both in maps (a), (b), and (c) refers to Indigenous and foreign languages combined. Knowledge of what is an Indigenous language is based exclusively on Ethnologue ( https://www.ethnologue.com ).
Ask Lune
Ask anything. The answer quotes the paper.
Lune reads all of it and the papers around it, then shows you the passages behind every answer.
Lune answers from
the full paper · 2 top-tier papers that cite it · related work
Free to start. No credit card required.
Cited by top-tier papers2
- AfriqueLLM: How Data Mixing and Model Architecture Impact Continued Pre-training for African LanguagesHao Yu, Tianyi Xu, Michael A. Hedderich, Wassim Hamidouche et al.ACL 2026 · 5 citations
- Voice of a Continent: Mapping Africa's Speech Technology FrontierAbdelRahim A. Elmadany, Sang Yun Kwon, Hawau Olamide Toyin, Alcides Alcoba Inciarte et al.EMNLP 2025
Builds on11
- XGLUE: A New Benchmark Datasetfor Cross-lingual Pre-training, Understanding and GenerationYaobo Liang, Nan Duan, Yeyun Gong, Ning Wu et al.EMNLP 2020 · 232 citations
- XL-Sum: Large-Scale Multilingual Abstractive Summarization for 44 LanguagesTahmid Hasan, Abhik Bhattacharjee, Md. Saiful Islam, Kazi Samin Mubasshir et al.ACL Findings 2021 · 228 citations
- KorNLI and KorSTS: New Benchmark Datasets for Korean Natural Language UnderstandingJiyeon Ham, Yo Joong Choe, Kyubyong Park, Ilji Choi et al.EMNLP Findings 2020 · 62 citations
- MasakhaNER 2.0: Africa-centric Transfer Learning for Named Entity RecognitionDavid Ifeoluwa Adelani, Graham Neubig, Sebastian Ruder, Shruti Rijhwani et al.EMNLP 2022 · 46 citations
- The State and Fate of Linguistic Diversity and Inclusion in the NLP WorldPratik Joshi, Sebastin Santy, Amar Budhiraja, Kalika Bali et al.ACL 2020 · 40 citations
Related papers
- AfriInstruct: Instruction Tuning of African Languages for Diverse TasksKosei Uemura, Mahe Chen, Alex Pejovic, Chika Maduabuchi et al.EMNLP Findings 2024 · 2 citations
- AfroBench: How Good are Large Language Models on African Languages?Jessica Ojo, Odunayo Ogundepo, Akintunde Oladipo, Kelechi Ogueji et al.ACL Findings 2025 · 16 citations
- SERENGETI: Massively Multilingual Language Models for AfricaIfe Adebara, AbdelRahim A. Elmadany, Muhammad Abdul-Mageed, Alcides Alcoba InciarteACL Findings 2023 · 18 citations
- Afri-MCQA: Multimodal Cultural Question Answering for African LanguagesAtnafu Lambebo Tonja, Srija Anand, Emilio Villa-Cueva, Israel Abebe Azime et al.ACL 2026 · 2 citations
- Charting the Landscape of African NLP: Mapping Progress and Shaping the Road AheadJesujoba Oluwadara Alabi, Michael A. Hedderich, David Ifeoluwa Adelani, Dietrich KlakowEMNLP 2025
