Saved in:
Bibliographic Details
Main Authors: Duong, Phan Anh, Luong, Cat, Bommana, Divyesh, Jiang, Tianyu
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.01047
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916968814084096
author Duong, Phan Anh
Luong, Cat
Bommana, Divyesh
Jiang, Tianyu
author_facet Duong, Phan Anh
Luong, Cat
Bommana, Divyesh
Jiang, Tianyu
contents Emotions manifest through physical experiences and bodily reactions, yet identifying such embodied emotions in text remains understudied. We present an embodied emotion classification dataset, CHEER-Ekman, extending the existing binary embodied emotion dataset with Ekman's six basic emotion categories. Using automatic best-worst scaling with large language models, we achieve performance superior to supervised approaches on our new dataset. Our investigation reveals that simplified prompting instructions and chain-of-thought reasoning significantly improve emotion recognition accuracy, enabling smaller models to achieve competitive performance with larger ones. Our dataset is publicly available at: https://github.com/menamerai/cheer-ekman.
format Preprint
id arxiv_https___arxiv_org_abs_2506_01047
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CHEER-Ekman: Fine-grained Embodied Emotion Classification
Duong, Phan Anh
Luong, Cat
Bommana, Divyesh
Jiang, Tianyu
Computation and Language
Emotions manifest through physical experiences and bodily reactions, yet identifying such embodied emotions in text remains understudied. We present an embodied emotion classification dataset, CHEER-Ekman, extending the existing binary embodied emotion dataset with Ekman's six basic emotion categories. Using automatic best-worst scaling with large language models, we achieve performance superior to supervised approaches on our new dataset. Our investigation reveals that simplified prompting instructions and chain-of-thought reasoning significantly improve emotion recognition accuracy, enabling smaller models to achieve competitive performance with larger ones. Our dataset is publicly available at: https://github.com/menamerai/cheer-ekman.
title CHEER-Ekman: Fine-grained Embodied Emotion Classification
topic Computation and Language
url https://arxiv.org/abs/2506.01047