Bots of a Feather: Mixing Biases in LLMs’ Opinion Dynamics
Published in International Conference on Complex Networks and Their Applications, 2024
This paper examines how biases mix and propagate in opinion dynamics when modeled through LLM-based agents. By simulating interactions between agents with different predispositions, the authors show how collective behaviors emerge and reveal mechanisms that shape polarization and consensus in artificial societies.
Recommended citation: Cau, E., Failla, A., & Rossetti, G. (2024, December). "Bots of a Feather: Mixing Biases in LLMs’ Opinion Dynamics." In International Conference on Complex Networks and Their Applications (pp. 166-176). Cham: Springer Nature Switzerland.
Download Paper | Download Slides