GPT-4 Emulates Average-Human Emotional Cognition from a Third-Person Perspective

• Published in: International Conference on Affective Computing and Intelligent Interaction and workshops pp. 337 - 345
• Main Authors: Tak, Ala N., Gratch, Jonathan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 15.09.2024
• Subjects: Accuracy; Affective computing; Annotations; Appraisal; appraisal theory; Cognition; Emotion recognition; GPT-4; Large language models; Neural activity; Safety

This paper extends recent investigations on the emotional reasoning abilities of Large Language Models (LLMs). Current research on LLMs has not directly evaluated the distinction between how LLMs predict the self-attribution of emotions and the perception of others' emotions. We first look at c...