Branching Narratives: Character Decision Points Detection

TL;DR

This paper introduces the CHADPOD task for identifying character decision points in narratives, presents a new dataset based on CYOA-like game graphs, and evaluates models achieving up to 89% accuracy, highlighting the complexity of narrative analysis.

Contribution

The paper proposes a novel task and dataset for detecting character decision points in narratives, and evaluates models demonstrating practical applications in story analysis.

Findings

Models achieved up to 89% accuracy on the task.

The dataset enables benchmarking of narrative decision point detection.

Identified challenges in understanding character-driven story dynamics.

Abstract

This paper presents the Character Decision Points Detection (CHADPOD) task, a task of identification of points within narratives where characters make decisions that may significantly influence the story's direction. We propose a novel dataset based on CYOA-like games graphs to be used as a benchmark for such a task. We provide a comparative analysis of different models' performance on this task, including a couple of LLMs and several MLMs as baselines, achieving up to 89% accuracy. This underscores the complexity of narrative analysis, showing the challenges associated with understanding character-driven story dynamics. Additionally, we show how such a model can be applied to the existing text to produce linear segments divided by potential branching points, demonstrating the practical application of our findings in narrative analysis.