Screenplay Summarization Using Latent Narrative Structure

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1920–1933 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 1920 Screenplay Summarization Using Latent Narrative Structure Pinelopi Papalampidi1 Frank Keller1 Lea Frermann2 Mirella Lapata1 1Institute for Language, Cognition and Computation School of Informatics, University of Edinburgh 2School of Computing and Information Systems University of Melbourne p.papalampidi@sms.ed.ac.uk, keller@inf.ed.ac.uk, lea.frermann@unimelb.edu.au, mlap@inf.ed.ac.uk Abstract Most general-purpose extractive summariza- tion models are trained on news articles, which are short and present all important information upfront. As a result, such models are biased by position and often perform a smart selection of sentences from the beginning of the doc- ument. When summarizing long narratives, which have complex structure and present in- formation piecemeal, simple position heuris- tics are not sufficient. In this paper, we pro- pose to explicitly incorporate the underlying structure of narratives into general unsuper- vised and supervised extractive summarization models. We formalize narrative structure in terms of key narrative events (turning points) and treat it as latent in order to summarize screenplays (i.e., extract an optimal sequence of scenes). Experimental results on the CSI corpus of TV screenplays, which we augment with scene-level summarization labels, show that latent turning points correlate with im- portant aspects of a CSI episode and improve summarization performance over general ex- tractive algorithms, leading to more complete and diverse summaries. 1 Introduction Automatic summarization has enjoyed renewed interest in recent years thanks to the popular- ity of modern neural network-based approaches (Cheng and Lapata, 2016; Nallapati et al., 2016, 2017; Zheng and Lapata, 2019) and the avail- ability of large-scale datasets containing hundreds of thousands of document–summary pairs (Sand- haus, 2008; Hermann et al., 2015; Grusky et al., 2018; Narayan et al., 2018; Fabbri et al., 2019; Liu and Lapata, 2019). Most efforts to date have con- centrated on the summarization of news articles which tend to be relatively short and formulaic following an “inverted pyramid” structure which places the most essential, novel and interesting el- Victim: Mike Kimble, found in a Body Farm. Died 6 hours ago, unknown cause of death. CSI discover cow tissue in Mike's body. Cross-contamination is suggested. Probable cause of death: Mike's house has been set on fire. CSI finds blood: Mike was murdered, fire was a cover up. First suspects: Mike's fiance, Jane and her ex-husband, Russ. CSI finds photos in Mike's house of Jane's daughter, Jodie, posing naked. Mike is now a suspect of abusing Jodie. Russ allows CSI to examine his gun. CSI discovers that the bullet that killed Mike was made of frozen beef that melt inside him. They also find beef in Russ' gun. Russ confesses that he knew that Mike was abusing Jody, so he confronted and killed him. CSI discovers that the naked photos were taken on a boat, which belongs to Russ. CSI discovers that it was Russ who was abusing his daughter based on fluids found in his sleeping bag and later killed Mike who tried to help Jodie. …

Content truncated to 500 words