Modeling Event Salience in Narratives via Barthes' Cardinal Functions
This work addresses the need for automated event salience estimation to aid tasks like story generation and text analysis in narratology and folkloristics, representing an incremental advancement in unsupervised NLP methods.
The paper tackled the problem of estimating event salience in narratives without annotations by adopting Barthes' definition and proposing unsupervised methods using pre-trained language models, showing that these methods outperform baselines on annotated folktales with fine-tuning on narrative texts as a key improvement factor.
Events in a narrative differ in salience: some are more important to the story than others. Estimating event salience is useful for tasks such as story generation, and as a tool for text analysis in narratology and folkloristics. To compute event salience without any annotations, we adopt Barthes' definition of event salience and propose several unsupervised methods that require only a pre-trained language model. Evaluating the proposed methods on folktales with event salience annotation, we show that the proposed methods outperform baseline methods and find fine-tuning a language model on narrative texts is a key factor in improving the proposed methods.