jiangg/chembert_cased

This is the pre-trained model presented in Automated Chemical Reaction Extraction from Scientific Literature, which is a BERT model trained on chemical literature data.

The training corpus was taken from ~200K ACS publications, more details can be found in the paper.

If using these models, please cite the following paper:

@article{guo2021automated,
  title={Automated Chemical Reaction Extraction from Scientific Literature},
  author={Guo, Jiang and Ibanez-Lopez, A Santiago and Gao, Hanyu and Quach, Victor and Coley, Connor W and Jensen, Klavs F and Barzilay, Regina},
  journal={Journal of Chemical Information and Modeling},
  year={2021},
  publisher={ACS Publications}
}