README.md

---
license: mit
---
# Model description 
LegalBert is a BERT-base-cased model fine-tuned on a subset of the `case.law` corpus. Further details can be found in this paper:

[A Dataset for Statutory Reasoning in Tax Law Entailment and Question Answering](http://ceur-ws.org/Vol-2645/paper5.pdf)  
Nils Holzenberger, Andrew Blair-Stanek and Benjamin Van Durme  
*Proceedings of the 2020 Natural Legal Language Processing (NLLP) Workshop, 24 August 2020*

# Usage
```
from transformers import AutoModel, AutoTokenizer
model = AutoModel.from_pretrained("jhu-clsp/LegalBert")
tokenizer = AutoTokenizer.from_pretrained("jhu-clsp/LegalBert")
```

# Citation
```
@inproceedings{holzenberger20dataset,
  author    = {Nils Holzenberger and
               Andrew Blair{-}Stanek and
               Benjamin Van Durme},
  title     = {A Dataset for Statutory Reasoning in Tax Law Entailment and Question
               Answering},
  booktitle = {Proceedings of the Natural Legal Language Processing Workshop 2020
               co-located with the 26th {ACM} {SIGKDD} International Conference on
               Knowledge Discovery {\&} Data Mining {(KDD} 2020), Virtual Workshop,
               August 24, 2020},
  series    = {{CEUR} Workshop Proceedings},
  volume    = {2645},
  pages     = {31--38},
  publisher = {CEUR-WS.org},
  year      = {2020},
  url       = {http://ceur-ws.org/Vol-2645/paper5.pdf},
}
```