🏢 University of British Columbia
Swan and ArabicMTEB: Dialect-Aware, Arabic-Centric, Cross-Lingual, and Cross-Cultural Embedding Models and Benchmarks
·4411 words·21 mins
AI Generated
🤗 Daily Papers
Natural Language Processing
Large Language Models
🏢 University of British Columbia
Swan & ArabicMTEB: New dialect-aware Arabic embedding models and benchmark achieve state-of-the-art performance, addressing limitations of existing multilingual models.