mteb/InjongoIntent

InjongoIntent An MTEB dataset Massive Text Embedding Benchmark Multicultural intent-classification dataset covering banking, home, kitchen & dining, travel and utility. 3 200 utterances per African language (2 240 / 320 / 640 train/dev/test) + 1 779 English. From ‘INJONGO: A Multicultural Intent Detection and Slot-filling Dataset for 16 African Languages’ (Yu et al., 2025). Task category t2c Domains Spoken Reference INJONGO: A Multicultural In…