Large Language Models (LLMs) are increasingly expected to navigate the nuances of human emotion. While research confirms that LLMs can simulate emotional intelligence, their internal emotional mechanisms remain largely unexplored. This paper investigates the latent emotional representations within modern LLMs by asking: how, where, and for how long is emotion encoded in their neural architecture? To address this, we introduce a novel, large-scale Reddit corpus of approximately 400,000 utterances, balanced across seven basic emotions through a multi-stage process of classification, rewriting, and synthetic generation. Using this dataset, we employ lightweight "probes" to read out information from the hidden layers of various Qwen3 and LLaMA models without altering their parameters. Our findings reveal that LLMs develop a surprisingly well-defined internal geometry of emotion, which sharpens with model scale and significantly outperforms zero-shot prompting. We demonstrate that this emotional signal is not a final-layer phenomenon but emerges early and peaks mid-network. Furthermore, the internal states are both malleable (they can be influenced by simple system prompts) and persistent, as the initial emotional tone remains detectable for hundreds of subsequent tokens. We contribute our dataset, an open-source probing toolkit, and a detailed map of the emotional landscape within LLMs, offering crucial insights for developing more transparent and aligned AI systems. The code and dataset are open-sourced.
AI with Emotions: Exploring Emotional Expressions in Large Language Models
Graduate School of Artificial Intelligence and Science, Rikkyo University(人工智能与科学研究生院,立命馆大学);Strategic Digital Business Unit, Mamezou Co., Ltd.(战略数字商务部,麦默株式会社)
Improved Emotional Alignment of AI and Humans: Human Ratings of Emotions Expressed by Stable Diffusion v1, DALL-E 2, and DALL-E 3
Delft University of Technology(代尔夫特理工大学);Delft Institute of Positive Design(代尔夫特积极设计研究所);Department of Human-Centered Design(以人为本设计系);Playpower Labs Pvt Ltd(Playpower实验室私人有限公司);School of Design(设计学院);Hong Kong Polytechnic University(香港理工大学);Department of Psychology(心理学系);Utrecht University(乌得勒支大学)
Emotion-Aware Prosodic Phrasing for Expressive Text-to-Speech
Inner Mongolia University(内蒙古大学);Shenzhen Research Institute of Big Data(深圳大数据研究院);The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳));National University of Singapore(新加坡国立大学)
Semantic Differentiation in Speech Emotion Recognition: Insights from Descriptive and Expressive Speech Roles
School of Electrical Engineering and Computer Science, University of Ottawa(电气工程与计算机科学学院,渥太华大学);School of Psychology, University of Ottawa(心理学学院,渥太华大学);National Research Council Canada, Ottawa, Canada(加拿大国家研究理事会,渥太华,加拿大)
Music Interpretation and Emotion Perception: A Computational and Neurophysiological Investigation
School of Electrical and Computer Engineering, National Technical University of Athens(电气与计算机工程学院,国家技术大学雅典);Department of Music Studies, National and Kapodistrian University of Athens(音乐研究系,国家与卡波迪斯特里亚大学雅典)
Controllable Expressive 3D Facial Animation via Diffusion in a Unified Multimodal Space
Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络安全学院);Laboratory for Big Data and Decision, School of Systems Engineering, National University of Defense Technology(国防科技大学系统工程学院大数据与决策实验室)