Citance-Contextualized Summarization of Scientific Papers

Shahbaz Syed, Ahmad Hakimi, Khalid Al-Khatib, Martin Potthast


Abstract
Current approaches to automatic summarization of scientific papers generate informative summaries in the form of abstracts. However, abstracts are not intended to show the relationship between a paper and the references cited in it. We propose a new contextualized summarization approach that can generate an informative summary conditioned on a given sentence containing the citation of a reference (a so-called “citance”). This summary outlines content of the cited paper relevant to the citation location. Thus, our approach extracts and models the citances of a paper, retrieves relevant passages from cited papers, and generates abstractive summaries tailored to each citance. We evaluate our approach using **Webis-Context-SciSumm-2023**, a new dataset containing 540K computer science papers and 4.6M citances therein.
Anthology ID:
2023.findings-emnlp.573
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2023
Month:
December
Year:
2023
Address:
Singapore
Editors:
Houda Bouamor, Juan Pino, Kalika Bali
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
8551–8568
Language:
URL:
https://aclanthology.org/2023.findings-emnlp.573
DOI:
10.18653/v1/2023.findings-emnlp.573
Bibkey:
Cite (ACL):
Shahbaz Syed, Ahmad Hakimi, Khalid Al-Khatib, and Martin Potthast. 2023. Citance-Contextualized Summarization of Scientific Papers. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 8551–8568, Singapore. Association for Computational Linguistics.
Cite (Informal):
Citance-Contextualized Summarization of Scientific Papers (Syed et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-emnlp.573.pdf
Video:
 https://aclanthology.org/2023.findings-emnlp.573.mp4