This paper presents RigoBERTa, a State-of-the-Art Language Model for Spanish. RigoBERTa is trained over a well-curated corpus formed up from different subcorpora with key features. It follows the DeBERTa architecture, which has several advantages over other architectures of similar size as BERT or RoBERTa. RigoBERTa performance is assessed over 13 NLU tasks in comparison with other available Spanish language models, namely, MarIA, BERTIN and BETO. RigoBERTa outperformed the three models in 10 out of the 13 tasks, achieving new "State-of-the-Art" results.
@article{arxiv.2205.10233,
title = {RigoBERTa: A State-of-the-Art Language Model For Spanish},
author = {Alejandro Vaca Serrano and Guillem Garcia Subies and Helena Montoro Zamorano and Nuria Aldama Garcia and Doaa Samy and David Betancur Sanchez and Antonio Moreno Sandoval and Marta Guerrero Nieto and Alvaro Barbero Jimenez},
journal= {arXiv preprint arXiv:2205.10233},
year = {2022}
}