Get To The Point: Summarization with Pointer-Generator Networks

TL;DR

This paper introduces a hybrid pointer-generator network with coverage mechanism for abstractive summarization, significantly improving factual accuracy and reducing repetition compared to previous models.

Contribution

It proposes a novel hybrid architecture combining copying and generation with coverage to enhance summarization quality.

Findings

Outperforms state-of-the-art on CNN/Daily Mail dataset by at least 2 ROUGE points.

Reduces repetition and improves factual accuracy in summaries.

Demonstrates effectiveness of combined pointer-generator and coverage mechanisms.

Abstract

Neural sequence-to-sequence models have provided a viable new approach for abstractive text summarization (meaning they are not restricted to simply selecting and rearranging passages from the original text). However, these models have two shortcomings: they are liable to reproduce factual details inaccurately, and they tend to repeat themselves. In this work we propose a novel architecture that augments the standard sequence-to-sequence attentional model in two orthogonal ways. First, we use a hybrid pointer-generator network that can copy words from the source text via pointing, which aids accurate reproduction of information, while retaining the ability to produce novel words through the generator. Second, we use coverage to keep track of what has been summarized, which discourages repetition. We apply our model to the CNN / Daily Mail summarization task, outperforming the current abstractive state-of-the-art by at least 2 ROUGE points.