Bert Small Mnli

🚀 Pytorch Pre-trained BERT Models

These Pytorch pre-trained models are converted from Tensorflow checkpoints from the official Google BERT repository. They were introduced in a specific paper and trained on MNLI.

🚀 Quick Start

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.

📚 Documentation

Citation

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

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

Performance

MNLI: 72.1%
MNLI-mm: 73.76%

Training Epochs

These models were trained for 4 epochs.

Author

@prajjwal_1