Copy the page URI to the clipboard
Gargett, Andrew; Firth, Rob and Aletras, Nikolaos
(2020).
DOI: https://doi.org/10.1109/BigData50022.2020.9378308
Abstract
We present a new, large-scale corpus for training and evaluating text summarization systems on legal opinions, called LegalOps. The corpus includes ~ 14K opinions together with their summaries from U.S. Federal Courts, e.g. the Supreme Court and Federal Appeals Courts. The aim of this paper is to provide a novel data source of sufficient variety that it will advance theoretical work on modeling the particular patterns within legal discourse, but also of sufficient size that it will provide a new challenging testbed for state-of-the-art automatic summarization models.