Scholarlink

In this paper, we leverage the capabilities of GPT-4o, which is integrated with cognitive modules such as memory, reflection, and chain of thought in simulating social network dynamics. Focusing on the Roe v. Wade case, which significantly influenced abortion laws and women’s rights in the United States, we model public opinion propagation on Twitter through likes, retweets, comments, and posts. Multiple simulation experiments are conducted to observe agents over different time intervals. By employing small-world networks and scale-free networks, we analyze the impact of various network structures. Our comparative study reveals variations in interaction patterns and potential echo chamber effects.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Simulating Social Network with LLM Agents: An Analysis of Information Propagation and Echo Chambers

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Wenzhen Zheng,
  • Xijin Tang

摘要

In this paper, we leverage the capabilities of GPT-4o, which is integrated with cognitive modules such as memory, reflection, and chain of thought in simulating social network dynamics. Focusing on the Roe v. Wade case, which significantly influenced abortion laws and women’s rights in the United States, we model public opinion propagation on Twitter through likes, retweets, comments, and posts. Multiple simulation experiments are conducted to observe agents over different time intervals. By employing small-world networks and scale-free networks, we analyze the impact of various network structures. Our comparative study reveals variations in interaction patterns and potential echo chamber effects.