Benchmarking-BERT-based-Models-for-Sentence-level-Topic-Classification-in-Nepali-Language Public Forked from bipeshrajsubedi/Benchmarking-BERT-based-Models-for-Sentence-level-Topic-Classification-in-Nepali-Language

This project benchmarks BERT-based models for sentence-level topic classification in Nepali, a low-resource language. Using 25K labeled sentences across five domains, it shows Indic models, especially MuRIL-large, outperform multilingual models, establishing a strong baseline for Nepali NLP research.