Contrastive Learning with Narrative Twins for Modeling Story Salience

TL;DR

This paper introduces a contrastive learning approach using narrative twins to model story salience, improving the identification of key events in narratives through specialized embeddings and operations.

Contribution

It proposes a novel contrastive learning framework with narrative twins for better story salience modeling, outperforming baseline methods on ROCStories and Wikipedia summaries.

Findings

Contrastively learned embeddings outperform masked-language-model baselines.

Summarization is the most reliable operation for identifying salient sentences.

Using narrative twins enhances the model's ability to distinguish important story elements.

Abstract

Understanding narratives requires identifying which events are most salient for a story's progression. We present a contrastive learning framework for modeling narrative salience that learns story embeddings from narrative twins: stories that share the same plot but differ in surface form. Our model is trained to distinguish a story from both its narrative twin and a distractor with similar surface features but different plot. Using the resulting embeddings, we evaluate four narratologically motivated operations for inferring salience (deletion, shifting, disruption, and summarization). Experiments on short narratives from the ROCStories corpus and longer Wikipedia plot summaries show that contrastively learned story embeddings outperform a masked-language-model baseline, and that summarization is the most reliable operation for identifying salient sentences. If narrative twins are not available, random dropout can be used to generate the twins from a single story. Effective distractors can be obtained either by prompting LLMs or, in long-form narratives, by using different parts of the same story.