# RoBERTa architecture optimization

RobBERT-2023 is the latest language model based on Dutch. It adopts the RoBERTa architecture and uses an updated Dutch tokenizer and training data to adapt to the latest development and changes in Dutch.

Pre-trained model based on RoBERTa architecture, specifically optimized for psychological counseling and therapy dialogue texts

BERTweet.BR is a language model specifically pre-trained for Portuguese tweets, based on the RoBERTa architecture and trained on 100 million Portuguese tweets.

Pre-trained language model optimized for Spanish biomedical and clinical texts, supporting clinical NLP tasks

Spanish tweet sentiment analysis model based on RoBERTuito, supporting POS(positive)/NEG(negative)/NEU(neutral) three-class sentiment classification

MagBERT-NER is a cutting-edge named entity recognition model specifically designed for Moroccan French (Maghreb region).

Greek language model based on RoBERTa, specifically designed for Greek social media text

A RoBERTa BASE model pre-trained on 78.5GB of Thai text, suitable for masked language modeling and text classification tasks

Catalan pre-trained language model based on RoBERTa architecture, developed by Spanish government agencies

Pretrained model based on RoBERTa architecture for Southern Quechua (Collao and Chanka dialects), trained with 5 million tokens