跳到正文
arXiv cs.MA· David Luo, Alberto Leon-Garcia·· 23 小时前AI 评分38

智能体原生应用:一种新的应用范式

Agent-Native Applications: A New Application Paradigm

AI 导读

论文提出“智能体原生应用”这一新范式,让基于 LLM 的智能体承担应用核心逻辑,并可在运行时调整应用图本身。为此作者给出架构模型:应用被表示为智能体、工具、数据源与 HITL 检查点的可移植图,允许行为以“行为包络”写入声明式应用规格;运行时由应用编排器作控制平面、智能体网格作数据平面并产出审计事件。

正文 · 原文

View PDF HTML (experimental)

Abstract:Traditional application architectures assume behavioral logic authored in advance, leaving reachable behavior largely bounded by explicit code and workflow rules. Large language model (LLM)-based agents challenge this assumption by enabling runtime reasoning and autonomous action to become part of application behavior, allowing applications to address situations not enumerated at design time. Extending this pattern, this paper identifies a new application paradigm in which LLM-based agents serve as central reasoning and action components responsible for the application's core logic and, where permitted, for adapting the application graph itself at runtime. We call these agent-native applications. While such applications significantly expand their possible behavioral space beyond explicit code and workflow rules, they also face a major control problem in which useful agentic reasoning should be preserved while application behavior should remain within a permissible space. We therefore propose an architectural model that represents the application as a portable graph of agents, tools, data sources, and human-in-the-loop (HITL) checkpoints, and encodes the application's permitted behavior as a behavioral envelope within a declarative application specification. At runtime, an application orchestrator serves as the control plane that coordinates tasks and governs how the graph and its permissions evolve, while an agent mesh serves as the data plane that mediates policy-relevant interactions and produces audit events. We then discuss the trust layer that makes agent-native applications governable and the supporting foundations required for practical operation. Two use cases illustrate the architecture, while Agent-Native Runtime (ANR) demonstrates selected core mechanisms in an executable prototype.
Comments: 21 pages, 8 figures, 1 table, 1 code listing; published in IEEE Access
Subjects: Software Engineering (cs.SE); Multiagent Systems (cs.MA)
Cite as: arXiv:2611.23243 [cs.SE]
  (or arXiv:2611.23243v1 [cs.SE] for this version)
  https://doi.org/10.48550/arXiv.2611.23243

arXiv-issued DOI via DataCite

Journal reference: IEEE Access, vol. 14, pp. 121508-121528, 2026
Related DOI: https://doi.org/10.1109/ACCESS.2026.3720448

DOI(s) linking to related resources

Submission history

From: David Luo [view email]
[v1] Thu, 3 Sep 2026 03:58:40 UTC (2,546 KB)

来源:arXiv cs.MA · arxiv.org