CHIRON: Rich Character Representations in Long-Form Narratives

TL;DR

CHIRON introduces a novel character representation method for long narratives, combining large language models and automated reasoning to improve character understanding and story analysis.

Contribution

It presents a new character sheet approach that organizes and filters character information, enhancing story analysis and generation capabilities.

Findings

CHIRON outperforms summary-based baselines in masked-character prediction.

Metrics from CHIRON align with human judgments of character-centricity.

The method improves character understanding in long-form narratives.

Abstract

Characters are integral to long-form narratives, but are poorly understood by existing story analysis and generation systems. While prior work has simplified characters via graph-based methods and brief character descriptions, we aim to better tackle the problem of representing complex characters by taking inspiration from advice given to professional writers. We propose CHIRON, a new `character sheet' based representation that organizes and filters textual information about characters. We construct CHIRON sheets in two steps: a Generation Module that prompts an LLM for character information via question-answering and a Validation Module that uses automated reasoning and a domain-specific entailment model to eliminate false facts about a character. We validate CHIRON via the downstream task of masked-character prediction, where our experiments show CHIRON is better and more flexible than comparable summary-based baselines. We also show that metrics derived from CHIRON can be used to automatically infer character-centricity in stories, and that these metrics align with human judgments.