多智能体回声室背后的社会电路
Social Circuits behind Multi-agent Echo Chambers
- Simon Fraser University（西蒙弗雷泽大学）
- University of Alberta（阿尔伯塔大学）
- Southern University of Science and Technology（南方科技大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对多智能体通信中的回声室问题，提出社会电路框架追踪消息对接收者激活的影响，并据此设计CGD方法选择有用消息，在三个模型四个数据集上取得最高或并列最高准确率且生成更少令牌。
AI中文摘要：
语言模型智能体通过交换消息来整合证据，但它们的通信也可能产生回声室效应，从而强化共享的错误。然而，整体任务性能并不能解释一条消息如何改变接收智能体的内部激活并影响其决策。在这项工作中，我们引入了社会电路（Social Circuits），这是一个通过接收者激活来追踪消息效应的框架。我们比较接收者在改变一条消息前后的答案。然后，我们恢复在原始消息下记录的选定激活，以确定这些激活在多大程度上重现了消息效应。基于社会电路，我们提出了电路引导的审议（Circuit-Guided Deliberation，CGD），该方法利用接收者激活变化来学习选择有用的消息。我们确定了激活替换在何种情况下能保留接收者的决策，并界定了CGD的任务性能与通过消息选择所能达到的最佳性能之间的差距。实验表明，接收者激活变化解释了消息效应，并指导消息选择，从而提升了任务性能。在三个模型和四个数据集上，CGD在我们的主要比较中取得了最高或并列最高的平均准确率，同时生成的令牌数少于多智能体基线。
英文摘要：
Language-model agents exchange messages to combine evidence, but their communication can also create echo chambers that reinforce shared errors. However, overall task performance does not explain how a message changes the receiving agent's internal activations and affects its decision. In this work, we introduce Social Circuits, a framework for tracing message effects through receiver activations. We compare the receiver's answers before and after changing a message. Then, we restore selected activations recorded under the original message to determine how much of the message effect these activations reproduce. Based on Social Circuits, we propose Circuit-Guided Deliberation (CGD), which learns to select useful messages using receiver activation changes. We establish when activation replacement preserves receiver decisions and bound the gap between CGD's task performance and the best achievable through message selection. Experiments show that receiver activation changes explain the message effects and guide message selection that improves the task performance. Across three models and four datasets, CGD achieves the highest or joint-highest average accuracy in our main comparisons while generating fewer tokens than multi-agent baselines.
