BERT (double)

Model and tokenizer files for BERT (double) model from When Does Pretraining Help? Assessing Self-Supervised Learning for Law and the CaseHOLD Dataset.

Training Data

BERT (double) is pretrained using the same English Wikipedia corpus that the base BERT model (uncased, 110M parameters), bert-base-uncased, was pretrained on. For more information on the pretraining corpus, refer to the BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding paper.

Training Objective

This model is initialized with the base BERT model (uncased, 110M parameters), bert-base-uncased, and trained for an additional 1M steps on the MLM and NSP objective.

This facilitates a direct comparison to our BERT-based models for the legal domain, which are also pretrained for 2M total steps.

Usage

Please see the casehold repository for scripts that support computing pretrain loss and finetuning on BERT (double) for classification and multiple choice tasks described in the paper: Overruling, Terms of Service, CaseHOLD.

See demo.ipynb in the casehold repository for details on calculating domain specificity (DS) scores for tasks or task examples by taking the difference in pretrain loss on BERT (double) and Legal-BERT. DS score may be readily extended to estimate domain specificity of tasks in other domains using BERT (double) and existing pretrained models (e.g., SciBERT).

Citation

@inproceedings{zhengguha2021,
    title={When Does Pretraining Help? Assessing Self-Supervised Learning for Law and the CaseHOLD Dataset},
    author={Lucia Zheng and Neel Guha and Brandon R. Anderson and Peter Henderson and Daniel E. Ho},
    year={2021},
    eprint={2104.08671},
    archivePrefix={arXiv},
    primaryClass={cs.CL},
    booktitle={Proceedings of the 18th International Conference on Artificial Intelligence and Law},
    publisher={Association for Computing Machinery}
}

Lucia Zheng, Neel Guha, Brandon R. Anderson, Peter Henderson, and Daniel E. Ho. 2021. When Does Pretraining Help? Assessing Self-Supervised Learning for Law and the CaseHOLD Dataset. In Proceedings of the 18th International Conference on Artificial Intelligence and Law (ICAIL '21), June 21-25, 2021, São Paulo, Brazil. ACM Inc., New York, NY, (in press). arXiv: 2104.08671 [cs.CL].

Downloads last month
40
Hosted inference API
Fill-Mask
Examples
Examples
Mask token: [MASK]
This model can be loaded on the Inference API on-demand.