10_coref_output.zip
收藏DataCite Commons2025-05-29 更新2025-04-16 收录
下载链接：
https://archive.data.jhu.edu/file.xhtml?persistentId=doi:10.7281/T1/QVAHMD/ABAJNK
下载链接
链接失效反馈官方服务：
资源简介：
This contains the output of running Xia et al. (2020)’s coref model on all 965,467 federal cases in the corpus. For those who have case.law Researcher Access, we are happy to share the full coref output, including the underlying tokens. But for this public version, only the coref spans for each file are available, since one can reconstruct the full corpus if one also has the underlying tokens. The tokenization was done using SpanBERT/spanbert-base-cased.
提供机构：
Johns Hopkins University Data Archive
创建时间：
2022-05-18
