SZTAKI-HLT/hubert-base-cc

The model can be used as any other (cased) BERT model . It has been tested on chunking and NER tasks and set a new state-of-the-art on the former . Details of the training data and procedure can be found in the PhD thesis linked below . The exact scores are based on the exact scores given by BertForTokenClassification . The full evaluation of the full model will appear in a future paper . If you use the model, please cite the following papers: "Introducing huBERT." In: XVII Magyar Számítógépes Nyelvészeti Konferencia (,