Large Language Models in Human Behavior Modeling and Decision-Making

The field of human behavior modeling and decision-making is witnessing significant advancements with the integration of Large Language Models (LLMs). Recent developments indicate a shift towards leveraging LLMs to simulate complex human behaviors, such as financial trading, travel behavior, and insurance purchasing decisions. These models are being designed to learn from and adapt to various data streams, enabling more realistic and accurate simulations. The use of dual-agent frameworks, multi-timescale rewards, and retrieval-augmented generation are some of the innovative approaches being explored to improve the performance of LLMs in these tasks. Noteworthy papers in this area include FinPos, which proposes a position-aware trading agent system, and Aligning LLM agents with human learning and adjustment behavior, which introduces a novel dual-agent framework for simulating human travel behavior. InsurAgent, a Large Language Model-Empowered Agent for simulating individual behavior in purchasing flood insurance, is also a notable contribution. Additionally, Prompting for Policy and A Multi-Agent Psychological Simulation System for Human Behavior Modeling demonstrate the potential of LLMs in macroeconomic forecasting and human behavior modeling, respectively. The Psychogeography of Imaginary Places extends psychogeographical practice into virtual and fictive spaces, inviting new forms of meaning-making and self-reflection.

Sources

FinPos: A Position-Aware Trading Agent System for Real Financial Markets

Aligning LLM agents with human learning and adjustment behavior: a dual agent approach

InsurAgent: A Large Language Model-Empowered Agent for Simulating Individual Behavior in Purchasing Flood Insurance

Prompting for Policy: Forecasting Macroeconomic Scenarios with Synthetic LLM Personas

A Multi-Agent Psychological Simulation System for Human Behavior Modeling

The Psychogeography of Imaginary Places

Built with on top of