Multi-agent frameworks offer a powerful approach to enhancing both interaction-driven decision-making and complex reasoning in LLMs. This talk explores two applications of multi-agent interaction. First, we leverage LLM-based role-playing to simulate dialogues between sales agents and customers with diverse personalities, enabling behavioral analysis that informs personalized marketing strategies. Second, we demonstrate how multi-agent collaboration improves mathematical reasoning, where a nudging agent suggests strategic solving directions to a policy agent, fostering deeper thought processes and enhancing problem-solving performance. By optimizing both interactive and cognitive dimensions, multi-agent approaches pave the way for more adaptive and intelligent systems.
Yun-Nung (Vivian) Chen is currently a professor in the Department of Computer Science & Information Engineering at National Taiwan University. She earned her Ph.D. degree from Carnegie Mellon University, where her research interests focus on spoken dialogue systems and natural language processing. She was recognized as the World’s Top 2% Scientists in her 2023 impact, the Taiwan Outstanding Young Women in Science and received Google Faculty Research Awards, Amazon AWS Machine Learning Research Awards, MOST Young Scholar Fellowship, and FAOS Young Scholar Innovation Award. Her team was selected to participate in the first Alexa Prize TaskBot Challenge in 2021. Prior to joining National Taiwan University, she worked in the Deep Learning Technology Center at Microsoft Research Redmond.