Large language models (LLMs) have demonstrated impressive capabilities in text generation and reasoning, but real-world applications increasingly demand agents that can engage in strategic, interactive decision-making—not merely produce coherent language.
In this talk, I will present recent works toward building persuasive, aligned, and strategically competent language agents, with applications spanning effective marketing, competitive debate, and interactive gameplay.
Fei Fang,
Carnegie Mellon University
These agents are developed by integrating LLMs with reinforcement learning and game-theoretic principles, enabling them to reason under uncertainty, adapt to dynamic interactions, and optimize for long-term objectives. I will also discuss broader challenges in a future where AI agents take on increasingly strategic roles, including the need for auditing mechanisms that promote transparency, accountability, and alignment with human values.
Fei Fang is an associate professor at the Software and Societal Systems Department in the School of Computer Science at Carnegie Mellon University. Before joining CMU, she was a postdoctoral fellow at Harvard University, hosted by David Parkes and Barbara Grosz. She received her Ph.D. from the Department of Computer Science at the University of Southern California advised by Milind Tambe (now at Harvard).