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 2603.26993cs.MAcs.LGmath.OCstat.ML
收藏

基于大语言模型的多智能体规划的可靠性极限

On the Reliability Limits of LLM-Based Multi-Agent Planning

MIT(麻省理工学院)

Ruicheng Ao, Siyang Gao, David Simchi-Levi

AI总结:

本文研究了基于大语言模型的多智能体规划的可靠性极限,通过建模为有限无环决策网络,展示了在无新外部信号时,分布式网络在决策理论上被集中贝叶斯决策者主导,并分析了通信和信息压缩带来的损失。

摘要:

本文技术备忘录研究了基于大语言模型的多智能体规划的可靠性极限作为委托决策问题。我们将基于大语言模型的多智能体架构建模为有限无环决策网络,在其中多个阶段处理共享的模型-上下文信息,通过具有有限容量的语言接口进行通信,并可能调用人类审核。我们证明,在没有新外部信号的情况下,任何委托网络在决策理论上都被具有相同信息访问权限的集中贝叶斯决策者主导。在共同证据情况下,这表明在有限通信预算下优化多智能体有向无环图可以重新表述为在共享信号上的预算约束随机实验。我们还表征了通信和信息压缩带来的损失。在适当的评分规则下,集中贝叶斯价值与通信后的价值之间的差距可以表示为预期后验发散性,这在对数损失下减少为条件互信息,在布里尔分数下减少为预期后验平方误差。这些结果表征了委托大语言模型规划的根本可靠性极限。使用大语言模型在受控问题集上的实验进一步验证了这些表征。

英文摘要:

LLM-based multi-agent planning delegates operational decisions to specialists that communicate through finite reports. In this research, we study its reliability limits when a coordinator can revisit source records before completing a resource-constrained plan. A useful report can support an immediate decision or direct later reading, so its value depends on how these stages communicate. We derive exact communication and access tradeoffs and prove that two report forms suffice for a broad class of allocation requirements, one favoring immediate resource identification and the other distinguishing absence from unnamed availability. For shared capacity, this reduction permits joint optimization of reporting roles and adaptive reads for a fixed number of source types with unequal workloads and reading costs. Identical specialists can optimally report different facts, and maximizing immediate completion can impose an unbounded recovery cost penalty even with full-record access. We also conduct numerical experiments to quantify the benefits of the proposed joint optimization and show how source access improves language-agent decisions that already use their reports optimally. In particular, an exhaustive evaluation of 207,632 report designs on small allocation instances recovers the predicted joint optima. Our analysis suggests that reliable agent communication should be evaluated by the complete decisions it supports and the cost of finishing its unresolved cases, instead of solely by report length, local prediction accuracy, or immediate task completion, as is often the case in current practice.

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。

↑