Researchers have introduced Mawqif-v2, an extended Arabic dataset designed to evaluate cross-target generalization in stance detection. This new dataset includes 996 manually annotated Arabic tweets from three distinct targets: Women Driving, E-Cars, and the Trimester System. It serves as a held-out evaluation set, complementing the original Mawqif dataset used for training. The paper also provides baseline results using various transformer models and large language models to enable reproducible research. AI
IMPACT Enhances evaluation capabilities for Arabic NLP models, particularly in understanding nuanced public opinion across different topics.
RANK_REASON The cluster contains a research paper detailing a new benchmark dataset for NLP tasks. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
