Abstractive Dialogue Summarization with Sentence-Gated Modeling Optimized by Dialogue Acts

TL;DR

This paper introduces a sentence-gated neural model that explicitly incorporates dialogue acts to improve abstractive dialogue summarization, demonstrating significant performance gains on the AMI meeting corpus.

Contribution

It proposes a novel sentence-gated mechanism to leverage dialogue acts in neural summarization models, enhancing the quality of dialogue summaries.

Findings

Significant improvement over baselines on AMI corpus

Dialogue acts provide valuable cues for summarization

Model outperforms previous state-of-the-art methods

Abstract

Neural abstractive summarization has been increasingly studied, where the prior work mainly focused on summarizing single-speaker documents (news, scientific publications, etc). In dialogues, there are different interactions between speakers, which are usually defined as dialogue acts. The interactive signals may provide informative cues for better summarizing dialogues. This paper proposes to explicitly leverage dialogue acts in a neural summarization model, where a sentence-gated mechanism is designed for modeling the relationship between dialogue acts and the summary. The experiments show that our proposed model significantly improves the abstractive summarization performance compared to the state-of-the-art baselines on AMI meeting corpus, demonstrating the usefulness of the interactive signal provided by dialogue acts.