MADLAD-400 3B Multilingual Translation Model
Massively multilingual 3B-parameter translation model covering 400 languages with broad African-language inclusion.
Search titles, authors, subjects and dates from across the DARE African Knowledge Repository.
85 resources found in the repository.
Massively multilingual 3B-parameter translation model covering 400 languages with broad African-language inclusion.
Open language model adapted to African languages via continued pretraining, released by McGill-NLP.
Text-embedding model adapted for African languages and evaluated on the AfriMTEB benchmark of 59 languages and 14 tasks.
Open language model adapted to African languages via continued pretraining, released under CC BY 4.0 by McGill-NLP.
Open language model adapted to African languages through continued pretraining, released by McGill-NLP.
Large multilingual instruction dataset collection assembled for the Aya project, aggregating open datasets across 101 languages.
Evaluation set of GSM8K translations into 16 African languages with 250 test problems per language for multilingual mathematical reasoning.
Largest publicly available topic-classification dataset based on Flores-200, covering 205 languages and dialects with a compact seven-class label set.
Community-maintained successor to FLORES-200 for multilingual MT evaluation with expanded language coverage.
Can trust norms within the African moral system support data gathering for Generative AI (GenAI) development in African society? Recent developments in the field of large language models, such as GenAI, including models like ChatGPT and Midjourney, have identified a common issue with these GenAI models known as “AI hal…
A technical AI safety course developed at the University of Cape Town, with online materials and distribution information on the course site.
Speech version of FLoRes with ~2,009 n-way parallel sentences per language read by native speakers, for ASR, ST and language ID evaluation.