# BNE pre-training

A Spanish named entity recognition (NER) model based on the RoBERTa architecture, pre-trained on the BNE corpus and fine-tuned on the CAPITEL dataset, with optimized performance for lowercase named entity recognition.

This is a RoBERTa large model optimized for Spanish question answering tasks, trained on a large corpus from the Spanish National Library

This model is a Spanish named entity recognition model based on the RoBERTa architecture, pre-trained on the large-scale BNE corpus and fine-tuned using the CAPITEL-NERC dataset

Spanish named entity recognition model based on RoBERTa architecture, pre-trained on BNE corpus and fine-tuned on CAPITEL-NERC dataset