Classify or Select: Neural Architectures for Extractive Document Summarization

Ramesh Nallapati, Bowen Zhou and Mingbo Ma,

Abstract:

We present two novel and contrasting Recurrent Neural Network (RNN) based architectures for extractive summarization of documents. The Classifier based architecture sequentially accepts or rejects each sentence in the original document order for its membership in the summary. The Selector architecture, on the other hand, is free to pick one sentence at a time in any arbitrary order to generate the extractive summary.

Our models under both architectures jointly capture the notions of salience and redundancy of sentences. In addition, these models have the advantage of being very interpretable, since they allow visualization of their predictions broken up by abstract features such as information content, salience and redundancy.

We show that our models reach or outperform state-of-the-art supervised models on two different corpora. We also recommend the conditions under which one architecture is superior to the other based on experimental evidence.

TL;DR: This paper presents two different neural architectures for extractive document summarization whose predictions are very interpretable, and show that they reach or outperform state-of-the-art supervised models.

Conflicts: us.ibm.com

Keywords: Natural language processing, Supervised Learning, Applications, Deep learning