Read Top News First: A Document Reordering Approach for Multi-Document News Summarization

Chao Zhao, Tenghao Huang, Somnath Basu Roy Chowdhury, Muthu Kumar Chandrasekaran, Kathleen McKeown, Snigdha Chaturvedi


Abstract
A common method for extractive multi-document news summarization is to re-formulate it as a single-document summarization problem by concatenating all documents as a single meta-document. However, this method neglects the relative importance of documents. We propose a simple approach to reorder the documents according to their relative importance before concatenating and summarizing them. The reordering makes the salient content easier to learn by the summarization model. Experiments show that our approach outperforms previous state-of-the-art methods with more complex architectures.
Anthology ID:
2022.findings-acl.51
Volume:
Findings of the Association for Computational Linguistics: ACL 2022
Month:
May
Year:
2022
Address:
Dublin, Ireland
Venues:
ACL | Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
613–621
Language:
URL:
https://aclanthology.org/2022.findings-acl.51
DOI:
10.18653/v1/2022.findings-acl.51
Bibkey:
Cite (ACL):
Chao Zhao, Tenghao Huang, Somnath Basu Roy Chowdhury, Muthu Kumar Chandrasekaran, Kathleen McKeown, and Snigdha Chaturvedi. 2022. Read Top News First: A Document Reordering Approach for Multi-Document News Summarization. In Findings of the Association for Computational Linguistics: ACL 2022, pages 613–621, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Read Top News First: A Document Reordering Approach for Multi-Document News Summarization (Zhao et al., Findings 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.findings-acl.51.pdf
Code
 zhaochaocs/mds-dr
Data
CNN/Daily MailMulti-News