SumREN: Summarizing Reported Speech about Events in News

Abstract

A primary objective of news articles is to establish the factual record for an event, frequently achieved by conveying both the details of the specified event (i.e., the 5 Ws; Who, What, Where, When and Why regarding the event) and how people reacted to it (i.e., reported statements). However, existing work on news summarization almost exclusively focuses on the event details. In this work, we propose the novel task of summarizing the reactions of different speakers, as expressed by their reported statements, to a given event. To this end, we create a new multi-document summarization benchmark, SumREN, comprising 745 summaries of reported statements from various public figures obtained from 633 news articles discussing 132 events. We propose an automatic silver-training data generation approach for our task, which helps smaller models like BART achieve GPT-3 level performance on this task. Finally, we introduce a pipeline-based framework for summarizing reported speech, which we empirically show to generate summaries that are more abstractive and factual than baseline query-focused summarization approaches.

Can LMs Generalize to Future Data? An Empirical Analysis on Text Summarization

“…The Multi-News dataset (Fabbri et al, 2019) is proposed for multi-document news summarization, in which the input is a set of news documents. The SumREN dataset (Reddy et al, 2023) is constructed for reported speech summarization that aims to summarize the reported statements made by a specific person in news documents. Other text summarization datasets focused on summarizing scientific articles (Cohan et al, 2018;Lu et al, 2020), legal documents (Kornilova and Eidelman, 2019), events (Ghalandari et al, 2020;Yoon et al, 2023;, or dialogues (Gliwa et al, 2019;Zhu et al, 2021;.…”