In our Affiliate Highlights series, we are sharing short interviews with members of our community, offering a glimpse into their research, interests, and what inspires their work. This year, we are randomly selecting assistant and associate professors from our Affiliates program.
Each feature includes a few brief questions (nothing too serious!) to help spark connections, highlight new opportunities, and maybe even introduce you to your next collaborator. This month, get to know Jessie Chin!
Jessie Chin | associate professor, School of Information Sciences
What are your main research interests?
My main research interests are to translate theories of cognitive and behavioral sciences or conduct use-inspired basic research to design and develop information experience and technology to augment human performance, such as self-regulated learning and health self-management. I am especially interested in designing information experiences in evolving technology with older adults. The recent studies focus on utilizing artificial intelligence as an interface to represent and make sense of the external world, to design and develop more effective, fruitful, and fun information experiences for people with diverse backgrounds.
What are you most excited about in your research this year?
Our team is very excited to leverage cognitive sciences and large language models (LLMs) to advance agent development, human-AI teaming, and understandings of agent and human behavior. There are three research themes we are working on this year. One theme is to improve the current LLMs to represent human minds and behavior, in which we improved LLMs and developed conversational agents to recognize individuals at different stages of behavioral change and provide appropriate motivational language for health promotion. The other theme is to develop agents that take advantage of LLM capacities to support human behavior, in which we have developed a multi-agent system with various personas to help the ideation and creative thinking of individuals. The last theme is to enhance the memory and reasoning mechanisms of LLM-based agents, in which we examine information and social behavior of multi-agent societies with different cognitive systems.