Adapting Pre-trained Language Models to African Languages via Multilingual Adaptive Fine-Tuning

Research

NLP methods

Docs live

This paper introduces multilingual adaptive fine-tuning (MAFT) applied to 17 of the most-resourced African languages, producing the AfroXLMR family of models. Removing non-African-script tokens cuts model size by roughly 50 percent while matching the accuracy of single-language adaptation on named entity recognition, topic classification and sentiment analysis.

Website

Category: Research
Pricing: Free / open
Country: 🌍 Pan-African
Last verified: 25 Aug 2026

{
  "name": "Adapting Pre-trained Language Models to African Languages via Multilingual Adaptive Fine-Tuning",
  "slug": "adapting-pre-trained-language-models-to-african-languages-via-multilingual-adaptive-fine-tuning",
  "category": "RESEARCH",
  "country": "Pan-African",
  "docs_status": "LIVE",
  "licensing_required": "NONE",
  "verified": false,
  "last_verified": "2026-08-25",
  "website": "https://arxiv.org/abs/2204.06487",
  "documentation_url": null
}

get_resource("adapting-pre-trained-language-models-to-african-languages-via-multilingual-adaptive-fine-tuning") — via the Wycord MCP server

Verification history

25 Aug 2026 · live
22 Aug 2026 · live
19 Aug 2026 · live
16 Aug 2026 · live
13 Aug 2026 · live
10 Aug 2026 · live
7 Aug 2026 · live
4 Aug 2026 · live

Automated checks run every few days. See all recent status changes

Verification history

Tags