A Generative Approach to Titling and Clustering Wikipedia Sections

Anjalie Field, Sascha Rothe, Simon Baumgartner, Cong Yu, Abe Ittycheriah


Abstract
We evaluate the performance of transformer encoders with various decoders for information organization through a new task: generation of section headings for Wikipedia articles. Our analysis shows that decoders containing attention mechanisms over the encoder output achieve high-scoring results by generating extractive text. In contrast, a decoder without attention better facilitates semantic encoding and can be used to generate section embeddings. We additionally introduce a new loss function, which further encourages the decoder to generate high-quality embeddings.
Anthology ID:
2020.ngt-1.9
Volume:
Proceedings of the Fourth Workshop on Neural Generation and Translation
Month:
July
Year:
2020
Address:
Online
Venues:
ACL | NGT | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
79–87
Language:
URL:
https://aclanthology.org/2020.ngt-1.9
DOI:
10.18653/v1/2020.ngt-1.9
Bibkey:
Cite (ACL):
Anjalie Field, Sascha Rothe, Simon Baumgartner, Cong Yu, and Abe Ittycheriah. 2020. A Generative Approach to Titling and Clustering Wikipedia Sections. In Proceedings of the Fourth Workshop on Neural Generation and Translation, pages 79–87, Online. Association for Computational Linguistics.
Cite (Informal):
A Generative Approach to Titling and Clustering Wikipedia Sections (Field et al., NGT 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.ngt-1.9.pdf
Video:
 http://slideslive.com/38929822