arXiv：cs.AI· Dani Roytburg, Daphne Ippolito·· 6 小时前AI 评分36
异构 LLM 模拟中区分基础模型与人格设定
Disentangling Models from Personas in Heterogeneous LLM Simulations
AI 导读
研究通过构建由多个不同基础模型驱动的异构社交网络模拟发现，智能体获得的互动量更多取决于其基础模型而非被赋予的人格设定。当混合中加入更多模型时，基础模型的吸引或排斥效应显著增强，暗示网络动态在规模化时可能收敛到基础模型效应。内容中介分析进一步显示了基础模型跨语境的可预测性，以及模型词汇模式与互动最大化风格之间的关系。
正文
Abstract:Multi-agent simulations with large language models (LLMs) often operate networks of agents with a single base model. This overlooks the inter-model effects which may dominate engagement dynamics in real-world deployments. To show this, we simulate a heterogeneous social network powered by several different base models and show that the amount of engagement an agent receives depends more on its base model than on its assigned persona. The attraction or repulsion effects of a base model strengthen dramatically when more models are added in the mix, suggesting that networks dynamics may converge to base model effects at scale. To help explain this effect, we conduct a series of content-mediating analyses, showing the predictability of base models across contexts as well as the relationship between a model's lexical patterns and an engagement-maximizing style. In light of recent developments in mass multi-agent interaction, this work underscores the relevance of heterogeneous compositions in driving the outcomes of those networks
| Comments: | Presented as a Spotlight Paper at the Second Workshop on Social Simulation with LLMS, Third Conference on Language Modeling, 2026 |
| Subjects: | Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Social and Information Networks (cs.SI) |
| Cite as: | arXiv:2610.07535 [cs.MA] |
| (or arXiv:2610.07535v1 [cs.MA] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07535 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Daniel Roytburg [view email]
[v1]
Tue, 6 Oct 2026 00:01:30 UTC (1,830 KB)
来源：arXiv：cs.AI · arxiv.org