Saved in:
Bibliographic Details
Main Authors: Luo, Meng, Li, Bobo, Xu, Shanqing, Zhang, Shize, Chen, Qiuchan, Han, Menglu, Chen, Wenhao, Huang, Yanxiang, Fei, Hao, Lee, Mong-Li, Hsu, Wynne
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.00971
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914361020252160
author Luo, Meng
Li, Bobo
Xu, Shanqing
Zhang, Shize
Chen, Qiuchan
Han, Menglu
Chen, Wenhao
Huang, Yanxiang
Fei, Hao
Lee, Mong-Li
Hsu, Wynne
author_facet Luo, Meng
Li, Bobo
Xu, Shanqing
Zhang, Shize
Chen, Qiuchan
Han, Menglu
Chen, Wenhao
Huang, Yanxiang
Fei, Hao
Lee, Mong-Li
Hsu, Wynne
contents Despite rapid progress in multimodal large language models (MLLMs), their capability for deep emotional understanding remains limited. We argue that genuine affective intelligence requires explicit modeling of Theory of Mind (ToM), the cognitive substrate from which emotions arise. To this end, we introduce HitEmotion, a ToM-grounded hierarchical benchmark that diagnoses capability breakpoints across increasing levels of cognitive depth. Second, we propose a ToM-guided reasoning chain that tracks mental states and calibrates cross-modal evidence to achieve faithful emotional reasoning. We further introduce TMPO, a reinforcement learning method that uses intermediate mental states as process-level supervision to guide and strengthen model reasoning. Extensive experiments show that HitEmotion exposes deep emotional reasoning deficits in state-of-the-art models, especially on cognitively demanding tasks. In evaluation, the ToM-guided reasoning chain and TMPO improve end-task accuracy and yield more faithful, more coherent rationales. In conclusion, our work provides the research community with a practical toolkit for evaluating and enhancing the cognition-based emotional understanding capabilities of MLLMs. Our dataset and code are available at: https://HitEmotion.github.io/.
format Preprint
id arxiv_https___arxiv_org_abs_2602_00971
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Unveiling the Cognitive Compass: Theory-of-Mind-Guided Multimodal Emotion Reasoning
Luo, Meng
Li, Bobo
Xu, Shanqing
Zhang, Shize
Chen, Qiuchan
Han, Menglu
Chen, Wenhao
Huang, Yanxiang
Fei, Hao
Lee, Mong-Li
Hsu, Wynne
Computer Vision and Pattern Recognition
Despite rapid progress in multimodal large language models (MLLMs), their capability for deep emotional understanding remains limited. We argue that genuine affective intelligence requires explicit modeling of Theory of Mind (ToM), the cognitive substrate from which emotions arise. To this end, we introduce HitEmotion, a ToM-grounded hierarchical benchmark that diagnoses capability breakpoints across increasing levels of cognitive depth. Second, we propose a ToM-guided reasoning chain that tracks mental states and calibrates cross-modal evidence to achieve faithful emotional reasoning. We further introduce TMPO, a reinforcement learning method that uses intermediate mental states as process-level supervision to guide and strengthen model reasoning. Extensive experiments show that HitEmotion exposes deep emotional reasoning deficits in state-of-the-art models, especially on cognitively demanding tasks. In evaluation, the ToM-guided reasoning chain and TMPO improve end-task accuracy and yield more faithful, more coherent rationales. In conclusion, our work provides the research community with a practical toolkit for evaluating and enhancing the cognition-based emotional understanding capabilities of MLLMs. Our dataset and code are available at: https://HitEmotion.github.io/.
title Unveiling the Cognitive Compass: Theory-of-Mind-Guided Multimodal Emotion Reasoning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2602.00971