Learning to Extract Coherent Summary via Deep Reinforcement Learning

TL;DR

This paper introduces a neural coherence model and a reinforcement learning-based summarizer that jointly optimize for coherence and informativeness, achieving state-of-the-art results in extractive summarization.

Contribution

It proposes a novel neural coherence model and a reinforcement learning framework for extractive summarization that enhances summary coherence without feature engineering.

Findings

The neural coherence model effectively captures cross-sentence coherence patterns.

The RNES model outperforms existing baselines on ROUGE metrics.

Summaries generated are more coherent and readable.

Abstract

Coherence plays a critical role in producing a high-quality summary from a document. In recent years, neural extractive summarization is becoming increasingly attractive. However, most of them ignore the coherence of summaries when extracting sentences. As an effort towards extracting coherent summaries, we propose a neural coherence model to capture the cross-sentence semantic and syntactic coherence patterns. The proposed neural coherence model obviates the need for feature engineering and can be trained in an end-to-end fashion using unlabeled data. Empirical results show that the proposed neural coherence model can efficiently capture the cross-sentence coherence patterns. Using the combined output of the neural coherence model and ROUGE package as the reward, we design a reinforcement learning method to train a proposed neural extractive summarizer which is named Reinforced Neural Extractive Summarization (RNES) model. The RNES model learns to optimize coherence and informative importance of the summary simultaneously. Experimental results show that the proposed RNES outperforms existing baselines and achieves state-of-the-art performance in term of ROUGE on CNN/Daily Mail dataset. The qualitative evaluation indicates that summaries produced by RNES are more coherent and readable.