Saved in:
Bibliographic Details
Main Authors: Suhara, Yoshi, Alikaniotis, Dimitris
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.04677
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910321143185408
author Suhara, Yoshi
Alikaniotis, Dimitris
author_facet Suhara, Yoshi
Alikaniotis, Dimitris
contents Neural abstractive summarization models make summaries in an end-to-end manner, and little is known about how the source information is actually converted into summaries. In this paper, we define input sentences that contain essential information in the generated summary as $\textit{source sentences}$ and study how abstractive summaries are made by analyzing the source sentences. To this end, we annotate source sentences for reference summaries and system summaries generated by PEGASUS on document-summary pairs sampled from the CNN/DailyMail and XSum datasets. We also formulate automatic source sentence detection and compare multiple methods to establish a strong baseline for the task. Experimental results show that the perplexity-based method performs well in highly abstractive settings, while similarity-based methods perform robustly in relatively extractive settings. Our code and data are available at https://github.com/suhara/sourcesum.
format Preprint
id arxiv_https___arxiv_org_abs_2402_04677
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Source Identification in Abstractive Summarization
Suhara, Yoshi
Alikaniotis, Dimitris
Computation and Language
Neural abstractive summarization models make summaries in an end-to-end manner, and little is known about how the source information is actually converted into summaries. In this paper, we define input sentences that contain essential information in the generated summary as $\textit{source sentences}$ and study how abstractive summaries are made by analyzing the source sentences. To this end, we annotate source sentences for reference summaries and system summaries generated by PEGASUS on document-summary pairs sampled from the CNN/DailyMail and XSum datasets. We also formulate automatic source sentence detection and compare multiple methods to establish a strong baseline for the task. Experimental results show that the perplexity-based method performs well in highly abstractive settings, while similarity-based methods perform robustly in relatively extractive settings. Our code and data are available at https://github.com/suhara/sourcesum.
title Source Identification in Abstractive Summarization
topic Computation and Language
url https://arxiv.org/abs/2402.04677