arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2610.19863cs.MA
收藏

当极端主义者进入大语言模型社会:研究多智能体系统中的极端行为

When Extremists Enter LLM Societies: Investigating Extreme Behavior in Multi-Agent Systems

College of Computer and Information Science, Southwest University(西南大学计算机与信息科学学院);Yuanbao Team, Tencent(腾讯元宝团队);School of Computer and Cyber Sciences, Communication University of China(中国传媒大学计算机与网络空间安全学院);State Key Laboratory of AI Safety, Institute of Computing Technology, CAS(中国科学院计算技术研究所人工智能安全国家重点实验室)

Guo Chen, Huaijin Xie, Ziwen Li, Reed Li, Yu Lu, Haibo Shi, Bingbing Xu, Junjie Huang

AI总结:

该研究针对LLM多智能体系统,提出EXAM极端主义智能体建模框架,发现小型极端尾部群体可引发依赖场景的集体转变,揭示了社会规律在LLM模拟中的语言介导机制。

摘要:

大语言模型(LLM)智能体日益作为计算主体应用于多智能体社会模拟。现有研究侧重代表性角色与平均情况行为,而社会科学表明,持续存在的尾部行为体对集体动态的影响不成比例。我们在受控的基于LLM的多智能体系统(LLM-MAS)中研究这一规律。在这些系统中,此类行为体不能简化为固定标量状态:提示词、自然语言历史及主干生成调节其行为与影响力。我们首先分析社会学中极端主义者的三个属性,并基于LLM原生视角对LLM极端主义者建模。我们引入EXAM,即极端主义智能体建模框架,该框架可针对特定任务配置极端主义智能体,并精确控制其群体比例、空间分布、交互协议及测量接口。在国会选举、金融危机、疫情舆情、迭代囚徒困境四个场景中,我们观察到小型持续尾部群体产生可测量的、依赖主干与场景的集体转变。EXAM揭示了这一社会科学规律如何通过基于LLM的社会模拟特有的语言介导机制显现。

英文摘要:

Large language model (LLM) agents increasingly serve as computational actors in multi-agent social simulations. While existing work emphasizes representative personas and average-case behavior, social science shows that persistent tail actors can disproportionately shape collective dynamics. We study this regularity in controlled LLM-based multi-agent systems (LLM-MAS). In these systems, such actors cannot be reduced to fixed scalar states: prompts, natural-language histories, and backbone generation mediate their behavior and influence. We first analyzed three properties of extremists in sociology and model the LLM extremist based on LLM-native perspective. We introduce EXAM, an EXtremist Agent Modeling framework that capable of configuring extremist agents for specific tasks and precisely controlling their population proportion, spatial distribution, interaction protocols, and measurement interfaces. Across four scenarios: congressional election, financial crisis, epidemic public opinion, and iterated prisoner's dilemma, we observe that small persistent tail groups produce measurable, backbone- and scenario-dependent collective shifts. EXAM shows how this social-science regularity manifests through language-mediated mechanisms specific to LLM-based social simulation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑