CharMoral: A Character Morality Dataset for Morally Dynamic Character Analysis in Long-Form Narratives

Suyoung Bae, Gunhee Cho, Yun-Gyung Cheong, Boyang Li


Abstract
This paper introduces CharMoral, a dataset designed to analyze the moral evolution of characters in long-form narratives. CharMoral, built from 1,337 movie synopses, includes annotations for character actions, context, and morality labels. To automatically construct CharMoral, we propose a four-stage framework, utilizing Large Language Models, to automatically classify actions as moral or immoral based on context. Human evaluations and various experiments confirm the framework’s effectiveness in moral reasoning tasks in multiple genres. Our code and the CharMoral dataset are publicly available at https://github.com/BaeSuyoung/CharMoral.
Anthology ID:
2025.coling-main.589
Volume:
Proceedings of the 31st International Conference on Computational Linguistics
Month:
January
Year:
2025
Address:
Abu Dhabi, UAE
Editors:
Owen Rambow, Leo Wanner, Marianna Apidianaki, Hend Al-Khalifa, Barbara Di Eugenio, Steven Schockaert
Venue:
COLING
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
8809–8818
Language:
URL:
https://aclanthology.org/2025.coling-main.589/
DOI:
Bibkey:
Cite (ACL):
Suyoung Bae, Gunhee Cho, Yun-Gyung Cheong, and Boyang Li. 2025. CharMoral: A Character Morality Dataset for Morally Dynamic Character Analysis in Long-Form Narratives. In Proceedings of the 31st International Conference on Computational Linguistics, pages 8809–8818, Abu Dhabi, UAE. Association for Computational Linguistics.
Cite (Informal):
CharMoral: A Character Morality Dataset for Morally Dynamic Character Analysis in Long-Form Narratives (Bae et al., COLING 2025)
Copy Citation:
PDF:
https://aclanthology.org/2025.coling-main.589.pdf