Niklaus, Joël; Matoshi, Veton; Stürmer, Matthias; Chalkidis, Ilias; Ho, Daniel E (3 June 2023). MultiLegalPile: A 689GB Multilingual Legal Corpus Cornell University 10.48550/arXiv.2306.02069
|
Text
2306.02069.pdf Available under License Creative Commons: Attribution (CC-BY). Download (1MB) | Preview |
Large, high-quality datasets are crucial for training Large Language Models (LLMs). However, so far, there are few datasets available for specialized critical domains such as law and the available ones are often only for the English language. We curate and release MULTILEGALPILE, a 689GB corpus in 24 languages from 17 jurisdictions. The MULTILEGALPILE corpus, which includes diverse legal data sources with varying licenses, allows for pretraining NLP models under fair use, with more permissive licenses for the Eurlex Resources and Legal mC4 subsets. We pretrain two RoBERTa models and one Longformer multilingually, and 24 monolingual models on each of the language-specific subsets and evaluate them on LEXTREME. Additionally, we evaluate the English and multilingual models on LexGLUE. Our multilingual models set a new SotA on LEXTREME and our English models on LexGLUE. We release the dataset, the trained models, and all of the code under the most open possible licenses.
Item Type: |
Working Paper |
|---|---|
Division/Institute: |
Business School > Institute for Public Sector Transformation > Data and Infrastructure Business School |
Name: |
Niklaus, Joël 0000-0002-2779-1653; Matoshi, Veton 0009-0002-6613-5701; Stürmer, Matthias 0000-0001-9038-4041; Chalkidis, Ilias and Ho, Daniel E |
Publisher: |
Cornell University |
Language: |
English |
Submitter: |
Safiya Verbruggen |
Date Deposited: |
25 Aug 2023 11:48 |
Last Modified: |
09 Oct 2023 21:46 |
Publisher DOI: |
10.48550/arXiv.2306.02069 |
Related URLs: |
|
ARBOR DOI: |
10.24451/arbor.19714 |
URI: |
https://arbor.bfh.ch/id/eprint/19714 |
Actions (login required)
 |
View Item |