Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.16720 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914113536393216 |
|---|---|
| author | Sang, Jitao Xiao, Jinlin Han, Jiarun Chen, Jilin Chen, Xiaoyi Wei, Shuyu Sun, Yongjie Wang, Yuhang |
| author_facet | Sang, Jitao Xiao, Jinlin Han, Jiarun Chen, Jilin Chen, Xiaoyi Wei, Shuyu Sun, Yongjie Wang, Yuhang |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_16720 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Beyond Pipelines: A Survey of the Paradigm Shift toward Model-Native Agentic AI Sang, Jitao Xiao, Jinlin Han, Jiarun Chen, Jilin Chen, Xiaoyi Wei, Shuyu Sun, Yongjie Wang, Yuhang Artificial Intelligence 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. |
| title | Beyond Pipelines: A Survey of the Paradigm Shift toward Model-Native Agentic AI |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2510.16720 |