A Romanian BERT model, initialized from bert-base-romanian-cased-v1 and pretrained on the MARCELL v2.0 corpus of legal documents for 24h with the principles following the paper by Peter Izsak, Moshe Berchansky, Omer Levy, How to Train BERT with an Academic Budget
See more in: https://github.com/senisioi/rolegal
@inproceedings{ceausu-nisioi-2022-identifying,
title = "Identifying Draft Bills Impacting Existing Legislation: a Case Study on {R}omanian",
author = "Ceausu, Corina and
Nisioi, Sergiu",
booktitle = "Proceedings of the Thirteenth Language Resources and Evaluation Conference",
month = jun,
year = "2022",
address = "Marseille, France",
publisher = "European Language Resources Association",
url = "https://aclanthology.org/2022.lrec-1.391",
pages = "3670--3674",
}