Tag
Proposes SAGE, a neuro-symbolic framework combining language models with cognitive models for pragmatic reasoning, demonstrated on three case studies including referential expression generation and implicatures.
A new preprint introduces auto-psych, a framework that uses AI agents to propose cognitive models and design/conduct human experiments, extending AI scientific automation to psychology research.
This paper proposes Equation-to-Behavior Prompting and reinforcement learning to guide large language models to simulate diverse human decision-making patterns in persuasion games, showing improved belief accuracy and training outcomes.