Comparison of Czech Transformers on Text Classification Tasks
This addresses the need for better language-specific models for Czech, which is under-represented in multilingual models, but it is incremental as it applies established methods to a specific language.
The authors tackled the problem of limited performance of multilingual Transformers for Czech by pre-training and releasing two monolingual Czech Transformers, achieving improved results on text classification tasks compared to existing models.
In this paper, we present our progress in pre-training monolingual Transformers for Czech and contribute to the research community by releasing our models for public. The need for such models emerged from our effort to employ Transformers in our language-specific tasks, but we found the performance of the published multilingual models to be very limited. Since the multilingual models are usually pre-trained from 100+ languages, most of low-resourced languages (including Czech) are under-represented in these models. At the same time, there is a huge amount of monolingual training data available in web archives like Common Crawl. We have pre-trained and publicly released two monolingual Czech Transformers and compared them with relevant public models, trained (at least partially) for Czech. The paper presents the Transformers pre-training procedure as well as a comparison of pre-trained models on text classification task from various domains.