MER 2025: When Affective Computing Meets Large Language Models

2025-04-28

AI Summary

This study addresses the limitations of conventional affective computing—namely, its reliance on predefined emotion categories and poor generalizability—by proposing a large language model (LLM)-driven generative paradigm for emotion understanding. Methodologically, we introduce the first LLM-adapted generative emotion modeling framework, integrating semi-supervised learning, multimodal (text, speech, vision) fusion, fine-grained label expansion, and interpretable attention mechanisms; we further pioneer a cross-task collaborative learning paradigm linking emotion and personality. Key contributions include: (1) releasing MERTools—an open-source toolkit; (2) establishing a unified evaluation protocol; and (3) introducing an open benchmark dataset covering four tracks: semi-supervised emotion recognition, fine-grained emotion analysis, multimodal interpretable prediction, and emotion-enhanced personality recognition. Experimental results demonstrate significant improvements in accuracy and cross-domain generalization across multiple metrics, advancing the practical deployment of generative affective AI.

Technology Category

• Cognitive Modeling & Cognitive Systems: Affective Computing
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Emotional Intelligence

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

MER2025 is the third year of our MER series of challenges, aiming to bring together researchers in the affective computing community to explore emerging trends and future directions in the field. Previously, MER2023 focused on multi-label learning, noise robustness, and semi-supervised learning, while MER2024 introduced a new track dedicated to open-vocabulary emotion recognition. This year, MER2025 centers on the theme"When Affective Computing Meets Large Language Models (LLMs)".We aim to shift the paradigm from traditional categorical frameworks reliant on predefined emotion taxonomies to LLM-driven generative methods, offering innovative solutions for more accurate and reliable emotion understanding. The challenge features four tracks: MER-SEMI focuses on fixed categorical emotion recognition enhanced by semi-supervised learning; MER-FG explores fine-grained emotions, expanding recognition from basic to nuanced emotional states; MER-DES incorporates multimodal cues (beyond emotion words) into predictions to enhance model interpretability; MER-PR investigates whether emotion prediction results can improve personality recognition performance. For the first three tracks, baseline code is available at MERTools, and datasets can be accessed via Hugging Face. For the last track, the dataset and baseline code are available on GitHub.

Problem

• Shifting from categorical frameworks to LLM-driven emotion understanding
• Enhancing emotion recognition with semi-supervised and fine-grained methods
• Investigating multimodal cues for interpretable affective computing

Innovation

• LLM-driven generative emotion understanding methods
• Multimodal cues enhance model interpretability
• Semi-supervised learning for emotion recognition