Schematic for summary selection, rating, editing
In joint work with the Allen Institute for AI, the Clearinghouse has contributed metadata and summaries for about 4500 federal cases to a new machine learning dataset, in an effort to train and refine Natural Language Processing (NLP) models that can generate automated summaries. You can read about the project in its first paper: Shen, Lo, Yu, Dahlberg, Schlanger & Downey, Multi-LexSum: Real-world Summaries of Civil Rights Lawsuits at Multiple Granularities, currently posted here. This special collection tags the in-sample cases.
See Searchable Results