University Medical Center Utrecht

Useful HF resources and fantastic contributors for Dutch NLP are

Individuals

• Pieter Delobelle, homepage and git
• Bram van Roy and homepage
• Robin Smits and git
• Janneke van de Zwaan and git
• Yeb Havinga and git
• Wietse de Vries and git
• François Remy, homepage and git
• Maarten Grootendorst, homepage and git
• Piek Vossen and git
• Eva Rombouts and git
• Joeran Bosma and git

Organisations

• University Medical Center Utrecht
• NLPtown and homepage
• doc2query
• LT3, language and translation technology team, University of Gent and homepage
• Textgain and homepage
• ML6, homepage and git
• CLiPS, homepage and git
• DTAI Research Group, KU Leuven, homepage and git
• GroNLP, homepage
• CLTL, homepage and git
• Nederlands Forensic Institute, homepage and git
• Integraal Kanker centrum Nederland (iKNL)
• Erasmus Medical Informatics

NLP Libraries relevant for (Dutch) clinical NLP:

• Clinlp

Encoder models

• RobBERT 2023
• BERTje
• BelabBERT
• MedRoBERTa.nl
• CardioBERTa.nl
• CardioDeBERTa.nl
• DRAGON-longformer-large-domain-specific
• DRAGON-longformer-base-domain-specific
• DRAGON-roberta-large-domain-specific
• DRAGON-roberta-base-domain-specific
• DRAGON-bert-base-domain-specific

Contrastive encoder models

• BioLord 2023-M Dutch

Decoder models

• GPT-2 on mC4, GPT-2 finetuned on Dutch
• GPT-neo on mC4
• GEITje (based on Mistral)
• Fietje (based on Phi-2), Zust_fietje
• J1

NTMs

• NLLB200
• UL2, en-nl, UL2, nl-en
• OPUS MT, en-nl, OPUS MT, nl-en, OPUS MT Healthcare, nl-en
• Llama 2 MT, nl-en

Datasets

• SoNaR
• COW
• mc4 cleaned
• TWnC
• Gigacorpus
• CulturaX
• FineWeb
• FineWeb 2