Toward Unifying Text Segmentation and Long Document Summarization

Abstract

Text segmentation is important for signaling a document's structure. Without segmenting a long document into topically coherent sections, it is difficult for readers to comprehend the text, let alone find important information. The problem is only exacerbated by a lack of segmentation in transcripts of audio/video recordings. In this paper, we explore the role that section segmentation plays in extractive summarization of written and spoken documents. Our approach learns robust sentence representations by performing summarization and segmentation simultaneously, which is further enhanced by an optimizationbased regularizer to promote selection of diverse summary sentences. We conduct experiments on multiple datasets ranging from scientific articles to spoken transcripts to evaluate the model's performance. Our findings suggest that the model can not only achieve state-of-the-art performance on publicly available benchmarks, but demonstrate better crossgenre transferability when equipped with text segmentation. We perform a series of analyses to quantify the impact of section segmentation on summarizing written and spoken documents of substantial length and complexity.

GreedyCAS: Unsupervised Scientific Abstract Segmentation with Normalized Mutual Information

“…A well-formulated abstract forms a scientific inference that extends from premises (e.g., shared knowledge, experimental evidence, or observation) to conclusions (e.g., suggestions, claims, Ripple et al (2012)). A splitting of an abstract into a conclusion segment and a premise segment can help readers better comprehend how conclusions are drawn (Bahadoran et al, 2020) and is of interest for downstream research tasks such as argument generation (Schiller et al, 2021), knowledge retrieval (Hua et al, 2019), opinion analysis (Hulpus et al, 2019), and text summarization (Cho et al, 2022). Many abstracts, especially those from the biomedical domain, are structured to help the reader extract the conclusions (e.g., abstracts follow the IMRaD format (Nair and Nair, 2014;Dernoncourt and Lee, 2017) or the CONSORT format (Hopewell et al, 2008)).…”