arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.12229cs.MA
收藏

语言模型团队作为分布式系统

Language Model Teams as Distributed Systems

Princeton University(普林斯顿大学);Massachusetts Institute of Technology(麻省理工学院);University of Cambridge(剑桥大学);New York University(纽约大学)

Elizabeth Mieczkowski, Katherine M. Collins, Ilia Sucholutsky, Natalia Vélez, Thomas L. Griffiths

AI总结:

本文提出将分布式系统作为原则性基础,用于创建和评估语言模型团队,揭示LLM团队与分布式计算中的共同挑战和优势。

摘要:

大型语言模型(LLMs)的能力日益增强,促使近期对LLM团队的兴趣增加。然而,尽管在大规模上部署LLM团队有所增加,我们缺乏一个原则性的框架来解决关键问题,例如团队何时是有帮助的,应使用多少代理,结构如何影响性能——以及团队是否比单个代理更好。而不是通过试错法设计和测试这些可能性,我们提出使用分布式系统作为创建和评估LLM团队的原则性基础。我们发现许多在分布式计算中研究的基本优势和挑战也出现在LLM团队中,突显了这两个研究领域交叉所带来的丰富实践见解。

英文摘要:

Large language models (LLMs) are growing increasingly capable, prompting recent interest in LLM teams. Yet, despite increased deployment of LLM teams at scale, we lack a principled framework for addressing key questions such as when a team is helpful, how many agents to use, how structure impacts performance -- and whether a team is better than a single agent. Rather than designing and testing these possibilities through trial-and-error, we propose using distributed systems as a principled foundation for creating and evaluating LLM teams. We find that many of the fundamental advantages and challenges studied in distributed computing also arise in LLM teams, highlighting the rich practical insights that can come from the cross-talk of these two fields of study.

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。

↑