Logo Lanfrica

XNLI Extension for African Languages

Domain:

natural language processing

Record type:

dataset
Creator:
ton
Host:
This dataset contains cross-lingual sentence pair data for Natural Language Inference (NLI) model training. It extends the XNLI corpus with English paired with the following African languages: Hausa (ha) Igbo (ig) Kikuyu (ki) Yoruba (yo) Kinyarwanda (rw) Luo (luo) Nigerian Pidgin (pcm) Total Examples: 24,539 Languages: 7 African languages + English