Niklaus, Joël; Matoshi, Veton; Rani, Pooja; Galassi, Andrea; Stürmer, Matthias; Chalkidis, Ilias (2023). Lextreme: A multi-lingual and multi-task benchmark for the legal domain Ithaca, NY: Cornell University 10.48550/arXiv.2301.13126
|
Text
2301.13126.pdf Available under License Creative Commons: Attribution-Noncommercial-Share Alike (CC-BY-NC-SA). Download (2MB) | Preview |
Lately, propelled by the phenomenal advances around the transformer architecture, the legal NLP field has enjoyed spectacular growth. To measure progress, well curated and challenging benchmarks are crucial. However, most benchmarks are English only and in legal NLP specifically there is no multilingual benchmark available yet. Additionally, many benchmarks are saturated, with the best models clearly outperforming the best humans and achieving near perfect scores. We survey the legal NLP literature and select 11 datasets covering 24 languages, creating LEXTREME. To provide a fair comparison, we propose two aggregate scores, one based on the datasets and one on the languages. The best baseline (XLM-R large) achieves both a dataset aggregate score a language aggregate score of 61.3. This indicates that LEXTREME is still very challenging and leaves ample room for improvement. To make it easy for researchers and practitioners to use, we release LEXTREME on huggingface together with all the code required to evaluate models and a public Weights and Biases project with all the runs.
Item Type: |
Working Paper |
---|---|
Division/Institute: |
Business School > Institute for Public Sector Transformation Business School > Institute for Public Sector Transformation > Data and Infrastructure Business School |
Name: |
Niklaus, Joël0000-0002-2779-1653; Matoshi, Veton0009-0002-6613-5701; Rani, Pooja; Galassi, Andrea; Stürmer, Matthias0000-0001-9038-4041 and Chalkidis, Ilias |
Publisher: |
Cornell University |
Language: |
English |
Submitter: |
Joël Niklaus |
Date Deposited: |
25 Aug 2023 09:33 |
Last Modified: |
09 Oct 2023 21:46 |
Publisher DOI: |
10.48550/arXiv.2301.13126 |
ARBOR DOI: |
10.24451/arbor.19688 |
URI: |
https://arbor.bfh.ch/id/eprint/19688 |