NarraSum: A Large-Scale Dataset for Abstractive Narrative Summarization

TL;DR

NarraSum is a large-scale dataset of 122,000 narrative documents and summaries from movies and TV episodes, designed to advance research in abstractive narrative summarization and natural language understanding.

Contribution

The paper introduces NarraSum, a novel large-scale dataset specifically for narrative summarization, addressing the challenge of understanding event causality and character behaviors.

Findings

Significant performance gap between humans and models on NarraSum

Dataset covers diverse genres from movies and TV episodes

Aims to promote future research in summarization and language understanding

Abstract

Narrative summarization aims to produce a distilled version of a narrative to describe its most salient events and characters. Summarizing a narrative is challenging as it requires an understanding of event causality and character behaviors. To encourage research in this direction, we propose NarraSum, a large-scale narrative summarization dataset. It contains 122K narrative documents, which are collected from plot descriptions of movies and TV episodes with diverse genres, and their corresponding abstractive summaries. Experiments show that there is a large performance gap between humans and the state-of-the-art summarization models on NarraSum. We hope that this dataset will promote future research in summarization, as well as broader studies of natural language understanding and generation. The dataset is available at https://github.com/zhaochaocs/narrasum.