研究多智能体协作何时有效，揭示其依赖性能与成本差异。
When Should We Orchestrate Multiple Agents?
- 构建现实约束下的协作框架，考虑推理成本与可用性限制。
- 理论证明：仅当智能体间存在性能或成本差异时协作才有效。
- 实证验证协作在选策略、任务外包等场景中提升效率。
多智能体协作策略常高估性能并低估协作成本。本文设计一个在现实条件（如推理成本、可用性约束）下协调智能体的框架。理论上证明，只有当智能体间存在性能或成本差异时，协作才有效。实验上，通过模拟环境验证了协作在选择智能体、解决社会学中的罗杰斯悖论学习策略问题，以及用户研究中的问答任务外包中具有实际帮助。
原文摘要 · Abstract (English)
Strategies for orchestrating the interactions between multiple agents, both human and artificial, can wildly overestimate performance and underestimate the cost of orchestration. We design a framework to orchestrate agents under realistic conditions, such as inference costs or availability constraints. We show theoretically that orchestration is only effective if there are performance or cost differentials between agents. We then empirically demonstrate how orchestration between multiple agents can be helpful for selecting agents in a simulated environment, picking a learning strategy in the infamous Rogers' Paradox from social science, and outsourcing tasks to other agents during a question-answer task in a user study.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。