MechBERT Models: Finetuned on SQuAD
Collection
MechBERT models fine-tuned for question answering tasks using the SQuAD version 1 and 2 datasets.
•
8 items
•
Updated
This model is fine-tuned on the SQuAD dataset for question answering tasks. It is based on the pre-trained PureMechBERT-uncased
model, developed by Cambridge Molecular Engineering designed for superior performance on downstream domain-specific NLP tasks.
See the associated paper for more details.
If you use this model, please cite:
@article{mechbert-kumar2025,
title={MechBERT: Language Models for Extracting Chemical and Property Relationships about Mechanical Stress and Strain},
author={Pankaj Kumar, Saurabh Kabra, Jacqueline M. Cole},
journal={Journal of Chemical Information and Modelling},
doi={10.1021/acs.jcim.4c00857},
year={2025}
}