RobeCzech: Czech RoBERTa, a monolingual contextualized language representation model
This provides a specialized tool for Czech NLP applications, though it is incremental as it adapts an existing method to a new language.
The authors tackled the problem of lacking a high-performance monolingual language model for Czech by developing RobeCzech, a RoBERTa-based model trained on Czech data, which outperformed existing multilingual and Czech-trained models and achieved state-of-the-art results in four out of five evaluated NLP tasks.
We present RobeCzech, a monolingual RoBERTa language representation model trained on Czech data. RoBERTa is a robustly optimized Transformer-based pretraining approach. We show that RobeCzech considerably outperforms equally-sized multilingual and Czech-trained contextualized language representation models, surpasses current state of the art in all five evaluated NLP tasks and reaches state-of-the-art results in four of them. The RobeCzech model is released publicly at https://hdl.handle.net/11234/1-3691 and https://huggingface.co/ufal/robeczech-base.