AER-LLM is a novel framework that enhances large language models' ability to recognize complex, ambiguous emotions in text—moving beyond simple emotion labeling to more human-like understanding.

• Introduces an ambiguity-aware emotion recognition approach that captures the nuanced, multi-dimensional nature of human emotions
• Leverages the reasoning capabilities of LLMs to identify emotional complexity rather than forcing single-label classifications
• Demonstrates superior performance in recognizing ambiguous emotional expressions compared to traditional methods
• Provides a framework for more emotionally intelligent conversational AI

This research is particularly valuable for medical applications, enabling mental health assessment tools, empathetic patient care systems, and therapeutic conversational agents that can recognize and respond to subtle emotional cues.

AER-LLM: Ambiguity-aware Emotion Recognition Leveraging Large Language Models