Searching for Effective Neural Extractive Summarization: What Works and What's Next

TL;DR

This paper investigates neural extractive summarization models to understand their success, explores architectural and training improvements, and achieves state-of-the-art results on CNN/DailyMail.

Contribution

It provides insights into what makes neural extractive summarization effective and proposes enhancements that lead to significant performance improvements.

Findings

Identifies key factors influencing model performance.

Proposes an effective framework that outperforms previous methods.

Achieves new state-of-the-art on CNN/DailyMail dataset.

Abstract

The recent years have seen remarkable success in the use of deep neural networks on text summarization. However, there is no clear understanding of \textit{why} they perform so well, or \textit{how} they might be improved. In this paper, we seek to better understand how neural extractive summarization systems could benefit from different types of model architectures, transferable knowledge and learning schemas. Additionally, we find an effective way to improve current frameworks and achieve the state-of-the-art result on CNN/DailyMail by a large margin based on our observations and analyses. Hopefully, our work could provide more clues for future research on extractive summarization.