多用户大语言模型智能体

摘要

大型语言模型（LLM）及基于LLM的智能体正日益作为规划与决策助手被广泛应用，然而现有系统大多隐式遵循单一主体交互范式——模型被设计为满足单一主导用户的目标，其指令被视为唯一权威来源与效用标准。但随着这些系统被整合到团队工作流和组织工具中，它们越来越多地需要同时服务多个用户，每个用户都具有不同的角色、偏好和权限级别，从而形成多用户、多主体场景，不可避免地引发目标冲突、信息不对称和隐私约束等问题。本研究首次对多用户LLM智能体展开系统性探索。我们首先将多用户与LLM智能体的交互形式化为多主体决策问题，即单个智能体需协调多个潜在利益冲突用户及其相关挑战。随后提出统一的多用户交互协议，并设计三种针对性压力测试场景，以评估现有LLM在指令遵循、隐私保护和协同协作方面的能力。实验结果表明系统性缺陷：前沿LLM在用户目标冲突时难以保持稳定的优先级排序，在多轮交互中隐私泄露风险递增，且在需要迭代信息收集的协同场景中出现效率瓶颈。

English

Large language models (LLMs) and LLM-based agents are increasingly deployed as assistants in planning and decision making, yet most existing systems are implicitly optimized for a single-principal interaction paradigm, in which the model is designed to satisfy the objectives of one dominant user whose instructions are treated as the sole source of authority and utility. However, as they are integrated into team workflows and organizational tools, they are increasingly required to serve multiple users simultaneously, each with distinct roles, preferences, and authority levels, leading to multi-user, multi-principal settings with unavoidable conflicts, information asymmetry, and privacy constraints. In this work, we present the first systematic study of multi-user LLM agents. We begin by formalizing multi-user interaction with LLM agents as a multi-principal decision problem, where a single agent must account for multiple users with potentially conflicting interests and associated challenges. We then introduce a unified multi-user interaction protocol and design three targeted stress-testing scenarios to evaluate current LLMs' capabilities in instruction following, privacy preservation, and coordination. Our results reveal systematic gaps: frontier LLMs frequently fail to maintain stable prioritization under conflicting user objectives, exhibit increasing privacy violations over multi-turn interactions, and suffer from efficiency bottlenecks when coordination requires iterative information gathering.

多用户大语言模型智能体

Multi-User Large Language Model Agents

摘要

Support