Framing Effects in Independent-Agent Large Language Models: A Cross-Family Behavioral Analysis

Zice Wang, Zhenyu Zhang

In many real-world applications, large language models (LLMs) operate as independent agents without interaction, thereby limiting coordination. In this setting, we examine how prom…