MBZUAI-Paris/

EgyptianWinoGrande

	Dataset Card for EgyptianWinoGrande (Arabic and Latin Script)




	
	Dataset Description




	
	Dataset Summary

WinoGrande (Egyptian Arabic) is a coreference resolution benchmark designed to test a model’s ability to resolve pronouns in ambiguous contexts. Each question has two candidate nouns and one target pronoun, translated into Egyptian Arabic.

Task Category: Multiple-choice question answering Task: Selecting the correct answer from a list of… See the full description on the dataset page: https://huggingface.co/datasets/MBZUAI-Paris/EgyptianWinoGrande.