# prajjwal1 /bert-small-mnli

The following model is a Pytorch pre-trained model obtained from converting Tensorflow checkpoint found in the official Google BERT repository. These BERT variants were introduced in the paper Well-Read Students Learn Better: On the Importance of Pre-training Compact Models. These models are trained on MNLI.

If you use the model, please consider citing the paper

@misc{bhargava2021generalization,
title={Generalization in NLI: Ways (Not) To Go Beyond Simple Heuristics},
author={Prajjwal Bhargava and Aleksandr Drozd and Anna Rogers},
year={2021},
eprint={2110.01518},
archivePrefix={arXiv},
primaryClass={cs.CL}
}


Original Implementation and more info can be found in this Github repository.

MNLI: 72.1%
MNLI-mm: 73.76%


These models were trained for 4 epochs.

@prajjwal_1