aitslab/biobert_bc5cdr_disease_v2

According to our evaluation on benchmark corpora, the v1 models performs better than v2. Therefore, the authors suggest using v1 models. However, these models can be used for comparison and benchmarking. More information can be found in our github repo and paper. Please cite the paper, if you use the model.

https://github.com/Aitslab/EasyNER

@article{ahmed2023easyner, title={EasyNER: A Customizable Easy-to-Use Pipeline for Deep Learning- and Dictionary-based Named Entity Recognition from Medical Text}, author={Rafsan Ahmed and Petter Berntsson and Alexander Skafte and Salma Kazemi Rashed and Marcus Klang and Adam Barvesten and Ola Olde and William Lindholm and Antton Lamarca Arrizabalaga and Pierre Nugues and Sonja Aits}, year={2023}, eprint={2304.07805}, archivePrefix={arXiv}, primaryClass={q-bio.QM} }