为LLM智能体可控舆论动力学引入贝叶斯信念层
原标题：Bayesian Belief Layer for Controllable Opinion Dynamics in LLM Agents
AI 摘要
研究者提出 Bayesian Chronicle Agents (BCA)，在 LLM 智能体的 persona 与语言生成之间加入一个显式贝叶斯信念层，将每个立场表示为概率，每听到一句话做一次贝叶斯更新，并用单一 prior-strength 参数控制固执程度。该参数可扫描出共识、持续分歧、坚定少数影响三种经典舆论动力学机制，并在四个 LLM 上验证了信念排序的可恢复性与模拟的可审计性。作者开源了代码、提示词和运行日志。
正文节选
Bayesian Belief Layer for Controllable Opinion Dynamics in LLM Agents Abstract LLM agents in social simulation revise their opinions implicitly, in context: how open an agent is to persuasion can neither be specified nor verified, and collective outcomes inherit the model’s training prior. We introduce Bayesian Chronicle Agents (BCA), a minimal belief layer separating what an agent believes from how it speaks. Each stance is a probability, updated by one Bayesian step per utterance heard. A sing