LLMs for Behavioral Science

Given the promise of large language models (LLMs) for understanding and predicting human behavior, it is crucial to develop a clear plan for tapping into this rich resource and make behavioral scientists aware of best practices and potential pitfalls in the use of such models. The objective of the LLMs for Behavioral Science Workshop is to provide a platform for researchers to discuss the successes, challenges, and future possibilities of using language models for behavioral science research, create a collaborative hub of researchers, and develop a unified vision for the use of language models in the behavioral sciences.

One goal of the workshop is to provide a training opportunity for early-career (doctoral, postdoctoral) researchers through attending lectures, discussions, and interacting with experts in the field. The workshop also includes a hands-on training component to take place on the first day.

The workshop will gather a core faculty consisting of leading researchers in the computational and behavioral sciences as well as a number of early-career researchers in these fields who will engage in active and structured discussions around specific themes as detailed in the workshop program. See the section on people for a full list of participants.