Fast Abstractive Summarization with Reinforce-Selected Sentence Rewriting

TL;DR

This paper introduces a fast, hierarchical abstractive summarization model that selects and rewrites salient sentences, achieving state-of-the-art results with significantly improved speed and abstractiveness.

Contribution

The paper presents a novel sentence-level policy gradient method for hierarchical sentence selection and rewriting, enabling faster inference and higher-quality summaries.

Findings

Achieves state-of-the-art performance on CNN/Daily Mail dataset.

Enables 10-20x faster inference speed.

Demonstrates better generalization on DUC-2002 dataset.

Abstract

Inspired by how humans summarize long documents, we propose an accurate and fast summarization model that first selects salient sentences and then rewrites them abstractively (i.e., compresses and paraphrases) to generate a concise overall summary. We use a novel sentence-level policy gradient method to bridge the non-differentiable computation between these two neural networks in a hierarchical way, while maintaining language fluency. Empirically, we achieve the new state-of-the-art on all metrics (including human evaluation) on the CNN/Daily Mail dataset, as well as significantly higher abstractiveness scores. Moreover, by first operating at the sentence-level and then the word-level, we enable parallel decoding of our neural generative model that results in substantially faster (10-20x) inference speed as well as 4x faster training convergence than previous long-paragraph encoder-decoder models. We also demonstrate the generalization of our model on the test-only DUC-2002 dataset, where we achieve higher scores than a state-of-the-art model.