SummScreen: A Dataset for Abstractive Screenplay Summarization

TL;DR

SummScreen is a new dataset for abstractive summarization of TV series transcripts, highlighting challenges in capturing plot details and character information, and evaluating various models including neural and nearest neighbor approaches.

Contribution

Introduces SummScreen, a challenging dataset for TV series summarization, along with entity-centric evaluation metrics and analysis of model performances.

Findings

Oracle extractive approach outperforms neural models on automatic metrics.

Neural models struggle to fully exploit input transcripts.

Models can generate unfaithful facts, indicating room for improvement.

Abstract

We introduce SummScreen, a summarization dataset comprised of pairs of TV series transcripts and human written recaps. The dataset provides a challenging testbed for abstractive summarization for several reasons. Plot details are often expressed indirectly in character dialogues and may be scattered across the entirety of the transcript. These details must be found and integrated to form the succinct plot descriptions in the recaps. Also, TV scripts contain content that does not directly pertain to the central plot but rather serves to develop characters or provide comic relief. This information is rarely contained in recaps. Since characters are fundamental to TV series, we also propose two entity-centric evaluation metrics. Empirically, we characterize the dataset by evaluating several methods, including neural models and those based on nearest neighbors. An oracle extractive approach outperforms all benchmarked models according to automatic metrics, showing that the neural models are unable to fully exploit the input transcripts. Human evaluation and qualitative analysis reveal that our non-oracle models are competitive with their oracle counterparts in terms of generating faithful plot events and can benefit from better content selectors. Both oracle and non-oracle models generate unfaithful facts, suggesting future research directions.