Long-span summarization via local attention and content selection


Type
Conference Object
Change log
Authors
Gales, MJF 
Abstract

Transformer-based models have achieved state-of-the-art results in a wide range of natural language processing (NLP) tasks including document summarization. Typically these systems are trained by fine-tuning a large pre-trained model to the target task. One issue with these transformer-based models is that they do not scale well in terms of memory and compute requirements as the input length grows. Thus, for long document summarization, it can be challenging to train or fine-tune these models. In this work, we exploit large pre-trained transformer-based models and address long-span dependencies in abstractive summarization using two methods: local self-attention; and explicit content selection. These approaches are compared on a range of network configurations. Experiments are carried out on standard long-span summarization tasks, including Spotify Podcast, arXiv, and PubMed datasets. We demonstrate that by combining these methods, we can achieve state-of-the-art results on all three tasks in the ROUGE scores. Moreover, without a large-scale GPU card, our approach can achieve comparable or better results than existing approaches.

Description
Keywords
cs.CL, cs.CL
Journal Title
ACL-IJCNLP 2021 - 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, Proceedings of the Conference
Conference Name
The Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL-IJCNLP 2021)
Journal ISSN
Volume Title
Publisher
Rights
All rights reserved
Sponsorship
Cambridge Assessment (Unknown)
1. ALTA institute, Cambridge Assessment English, University of Cambridge 2. Cambridge International & St John’s College Scholarship