A Human-AI Interaction Framework for Dynamically Adaptive Agent Personality
**作者单位(英文)**
1. College of Computer Science and Technology, Zhejiang University, Hangzhou 310028, China
2. Innovation Center of Yangtze River Delta, Zhejiang University, Jiaxing 314100, China
3. Department of Psychology and Behavioral Sciences, Zhejiang University, Hangzhou 310028, China
4. School of Mechanical Engineering, Northwestern Polytechnical University, Xi'an 710072, China
Abstract:Advances in artificial intelligence have introduced new challenges for human-AI interaction (HAII), creating an urgent need for interaction approaches that are more natural and sustainable. Three major challenges of personality-driven interaction in the intelligent era are analyzed, and it is argued that HAII should ultimately move toward human-like interaction. Grounded in prior work in personality psychology, a dynamic HAII framework for personality-adaptive agents is developed, and its potential to improve interaction outcomes and address emerging challenges in intelligent systems is articulated. By examining the relationship between the proposed framework and traditional task-oriented human-computer interaction paradigms, new directions are suggested for theoretical research and practical applications toward next-generation personality-enabled agents.