LexGLUE
收藏资源简介:
LexGLUE是一个专为法律文本理解设计的基准数据集,由哥本哈根大学等机构创建。该数据集包含7个子数据集,涵盖了从欧洲人权法院到美国最高法院的多种法律文本,旨在评估和推动自然语言处理技术在法律领域的应用。数据集内容丰富,包括法律判决预测、信息提取、案例摘要和法律问题回答等多个任务,旨在通过标准化评估提升法律NLP模型的泛化能力和性能。
LexGLUE is a benchmark dataset specifically designed for legal text understanding, developed by institutions such as the University of Copenhagen and other relevant organizations. This dataset includes 7 sub-datasets, covering a wide spectrum of legal texts ranging from the European Court of Human Rights to the Supreme Court of the United States, with the core objective of evaluating and advancing the application of natural language processing (NLP) technologies in the legal domain. The dataset encompasses diverse tasks including legal judgment prediction, information extraction, case summarization, and legal question answering, and is intended to improve the generalization capability and performance of legal NLP models via standardized evaluation.

- 1LexGLUE: A Benchmark Dataset for Legal Language Understanding in English哥本哈根大学, 丹麦 · 2022年



