超越流水线：迈向模型原生智能体的范式转变综述

Beyond Pipelines: A Survey of the Paradigm Shift toward Model-Native Agentic AI

October 19, 2025

作者: Jitao Sang, Jinlin Xiao, Jiarun Han, Jilin Chen, Xiaoyi Chen, Shuyu Wei, Yongjie Sun, Yuhang Wang

cs.AI

摘要

智能体AI的快速发展标志着人工智能进入了一个新阶段，大型语言模型（LLMs）不再仅仅是响应，而是能够行动、推理和适应。本综述追溯了构建智能体AI的范式转变：从基于流水线的系统——其中规划、工具使用和记忆由外部逻辑编排，到新兴的模型原生范式——这些能力被内化于模型参数之中。我们首先将强化学习（RL）定位为实现这一范式转变的算法引擎。通过将学习从模仿静态数据重新定义为结果驱动的探索，RL支撑了跨语言、视觉和具身领域的LLM + RL + 任务的统一解决方案。在此基础上，本综述系统回顾了每种能力——规划、工具使用和记忆——如何从外部脚本模块演变为端到端学习的行为。此外，它还探讨了这一范式转变如何重塑了主要的智能体应用，特别是强调长期推理的深度研究智能体和强调具身交互的GUI智能体。最后，我们讨论了智能体能力的持续内化，如多智能体协作和反思，以及未来智能体AI中系统和模型层角色的演变。这些发展共同勾勒出模型原生智能体AI作为一个集成学习和交互框架的清晰轨迹，标志着从构建应用智能的系统向开发通过经验增长智能的模型的转变。

English

The rapid evolution of agentic AI marks a new phase in artificial intelligence, where Large Language Models (LLMs) no longer merely respond but act, reason, and adapt. This survey traces the paradigm shift in building agentic AI: from Pipeline-based systems, where planning, tool use, and memory are orchestrated by external logic, to the emerging Model-native paradigm, where these capabilities are internalized within the model's parameters. We first position Reinforcement Learning (RL) as the algorithmic engine enabling this paradigm shift. By reframing learning from imitating static data to outcome-driven exploration, RL underpins a unified solution of LLM + RL + Task across language, vision and embodied domains. Building on this, the survey systematically reviews how each capability -- Planning, Tool use, and Memory -- has evolved from externally scripted modules to end-to-end learned behaviors. Furthermore, it examines how this paradigm shift has reshaped major agent applications, specifically the Deep Research agent emphasizing long-horizon reasoning and the GUI agent emphasizing embodied interaction. We conclude by discussing the continued internalization of agentic capabilities like Multi-agent collaboration and Reflection, alongside the evolving roles of the system and model layers in future agentic AI. Together, these developments outline a coherent trajectory toward model-native agentic AI as an integrated learning and interaction framework, marking the transition from constructing systems that apply intelligence to developing models that grow intelligence through experience.