For what purpose was the dataset created?Was therea specific task in mind? Was there aspecific gap that needed to be filled? Please provide a description.The Multi-LexSum dataset was curated to facilitate the development of automaticsummarization methods for civil rights lawsuits.Recent advances in document summarization have led to impressive results in generating ashort description for passages typically in hundreds of words. However, the source inputs forsummarizing civil right lawsuits are considerably longer: they can contain up to 70k words onaverage.

With the advent of large language models, methods for abstractive summarization have made great strides, creating potential for use in applications to aid knowledge workers processing unwieldy document collections. One such setting is the Civil Rights Litigation Clearinghouse (CRLC, https://clearinghouse.net),

With the advent of large language models, methods for abstractive summarization have made great strides, creating potential for use in applications to aid knowledge workers processing unwieldy document collections. One such setting is the Civil Rights Litigation Clearinghouse (CRLC, https://clearinghouse.net), Today, summarization in the CRLC requires extensive training of lawyers and law students who spend hours per case understanding multiple relevant documents in order to produce high-quality summaries of key events and outcomes. Motivated by this ongoing real-world summarization effort, we introduce Multi-LexSum, a collection of 9,280 expert-authored summaries drawn from ongoing CRLC writing. Multi-LexSum presents a challenging multi-document summarization task given the length of the source documents, often exceeding two hundred pages per case.