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 2602.11865cs.AI
收藏

智能AI委派

Intelligent AI Delegation

Google DeepMind(谷歌DeepMind)

Nenad Tomašev, Matija Franklin, Simon Osindero

AI总结:

本文提出了一种适应性框架,用于智能AI委派,通过任务分配、授权、责任转移和信任机制,提升复杂环境下的任务分解与委派能力。

摘要:

AI代理能够应对日益复杂的任务。为了实现更宏伟的目标,AI代理需要能够有意义地将问题分解成可管理的子组件,并安全地将完成任务委托给其他AI代理和人类。然而,现有的任务分解和委派方法依赖于简单的启发式方法,无法动态适应环境变化,也无法稳健地处理意外故障。在这里,我们提出了一种适应性框架,用于智能AI委派——涉及任务分配的一系列决策,同时包含授权、责任、问责、角色和边界明确的规范、意图的清晰性,以及建立双方(或更多方)之间信任的机制。所提出的框架适用于复杂委派网络中的人类和AI委派者及被委派者,旨在为新兴的代理网络中的协议开发提供指导。

英文摘要:

AI agents are able to tackle increasingly complex tasks. To achieve more ambitious goals, AI agents need to be able to meaningfully decompose problems into manageable sub-components, and safely delegate their completion across to other AI agents and humans alike. Yet, existing task decomposition and delegation methods rely on simple heuristics, and are not able to dynamically adapt to environmental changes and robustly handle unexpected failures. Here we propose an adaptive framework for intelligent AI delegation - a sequence of decisions involving task allocation, that also incorporates transfer of authority, responsibility, accountability, clear specifications regarding roles and boundaries, clarity of intent, and mechanisms for establishing trust between the two (or more) parties. The proposed framework is applicable to both human and AI delegators and delegatees in complex delegation networks, aiming to inform the development of protocols in the emerging agentic web.

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。

↑