PersonaFuse: A Personality Activation-Driven Framework for Enhancing Human-LLM Interactions
COM2 Level 4
Executive Classroom, COM2-04-02
Abstract:
Recent advancements in Large Language Models (LLMs) demonstrate remarkable capabilities across various fields. These developments have led to more direct communication between humans and LLMs in various situations, such as social companionship and psychological support. However, LLMs often exhibit limitations in emotional perception and social competence during real-world conversations. These limitations partly originate from their inability to adapt their communication style and emotional expression to different social and task contexts. In this work, we introduce PersonaFuse, a novel LLM post-training framework that enables LLMs to adapt and express different personalities for varying situations. Inspired by Trait Activation Theory and the Big-Five personality model, PersonaFuse employs a Mixture-of-Expert architecture that combines persona adapters with a dynamic routing network, enabling contextual trait expression. Experimental results show that PersonaFuse substantially outperforms baseline models across multiple dimensions of social-emotional intelligence. Importantly, these gains are achieved without sacrificing general reasoning ability or model safety, which remain common limitations of direct prompting and supervised fine-tuning approaches. PersonaFuse also delivers consistent improvements in downstream human-centered applications, such as mental health counseling and review-based customer service. Finally, human preference evaluations against leading LLMs, including GPT-4o and DeepSeek, demonstrate that PersonaFuse achieves competitive response quality despite its comparatively smaller model size. These findings demonstrate that PersonaFuse offers a theoretically grounded and practical approach for developing social-emotional enhanced LLMs, marking a significant advancement toward more human-centric AI systems.
Bio:
Yi Yang is an Associate Professor and Lee Heng Fellow in the Department of Information Systems, Business Statistics and Operations Management at the Hong Kong University of Science and Technology (HKUST). He is also the Director of the Center for Business and Social Analytics (CBSA). He received his Ph.D. in Computer Science from Northwestern University. His research focuses on designing machine learning methods to tackle complex challenges in business and FinTech. His work has appeared in leading journals in the business domain, including Information Systems Research, Management Information Systems Quarterly, Journal of Marketing, Contemporary Accounting Research, and INFORMS Journal on Computing. He has also published in premier venues in machine learning and natural language processing, such as ACL, EMNLP, KDD, ICLR, TKDE, and TOIS.
He serves as an Associate Editor for the INFORMS Journal on Computing and the MIS Quarterly Special Issue on "The Institutional Press in the Digital Age." He is also a member of the Editorial Review Board of Information Systems Research, and serves as Guest Associate Editor for ISR and MIS Quarterly. He is also a Senior Area Chair for the ACL Rolling Review (ARR). His research on AI and finance has had significant industry impact. His work has been referenced in policy statements by the Hong Kong Government and endorsed by the Hong Kong Monetary Authority (HKMA). He is currently leading research collaborations with Ping An Insurance, HSBC, WeBank, and HKMA.