Multiple model architectures were pretrained on Dutch clinical reports using a masked language modeling (MLM) objective. The architectures are BERT base, RoBERTa base, RoBERTa large, Longformer base, and Longformer large. These were pretrained with three pretraining strategies: general-domain pretraining, domain-specific pretraining, or mixed-domain pretraining. Please check out the paper [PENDING] and the model cards below for more information.