LexGLUE: A Benchmark Dataset for Legal Language Understanding in English
ACL 2022, Volume 1: Long Papers
What it adds
Unifies seven English datasets with accessible evaluation code and comparable metrics.
Read with this limitation
English-only and simplified tasks; independent expert re-annotation and human-performance estimates are missing.
- Jurisdiction tags
- United States · European Union · Council of Europe
- Language
- English
Follow the source
Selected statements located in the linked primary sources. These results have not been independently reproduced here.
LexGLUE includes seven datasets spanning classification and multiple-choice legal NLU tasks.
3.2 Tasks and DatasetsLegal-BERT’s harmonic macro-F1 is 70.8 versus BERT’s 68.2; no model wins every task.
5.2 Experimental Results
Connected research
Cited by
LawBench: Benchmarking Legal Knowledge of Large Language Models
Positions Chinese tasks against English legal NLP benchmarks.
Inspect the sourceCited by
LEXTREME: A Multi-Lingual and Multi-Task Benchmark for the Legal Domain
Positions multilingual evaluation alongside LexGLUE.
Inspect the sourceEntry status, source edition & rights
- Index entry
- Source-linked summary, prepared with agent assistance. Independent scholarly review is pending.
- Archive admission
- This bibliographic entry does not confer admission to the versioned archive.
- Edition inspected
- 10.18653/v1/2022.acl-long.297
- Reading coverage
- Canonical PDF sections 1-7, ethics, Appendix A access instructions and relevant reference entries; detailed dataset design and result tables inspected.
- Rights
- External references only. Papers and research artifacts remain with their original hosts; their own terms apply.
- Entry updated
- 2026-10-15
Source-linked summary · independent assessment pending.