arXiv cs.AI· Junan Hu, Jian Liu, Jingxiang Lai, Jiarui Hu, Yiwei Sheng, Shuang Chen, Jian Li, Dazhao Du, Song Guo·· 11 小时前AI 评分35
GUI 智能体中的强化学习综述：迈向数字居民
Reinforcement Learning in GUI Agents: A Survey Toward Digital Inhabitants
AI 导读
一篇发表于 TMLR 的综述首次系统梳理了强化学习与 GUI 智能体的交叉领域，将现有方法归纳为离线 RL、在线 RL 与混合策略三类。分析指出，可靠性与可扩展性的矛盾正推动复合多层奖励架构的采用，GUI I/O 延迟瓶颈加速了向世界模型训练的转移，且 System-2 式审慎推理会自发涌现。作者据此提出覆盖过程奖励、持续 RL、认知架构与安全部署的路线图。
正文 · 原文
Abstract:Graphical User Interface (GUI) agents have emerged as a promising paradigm for intelligent systems that perceive and interact with graphical interfaces visually. Yet supervised fine-tuning alone cannot handle long-horizon credit assignment, distribution shifts, and safe exploration in irreversible environments, making Reinforcement Learning (RL) a central methodology for advancing automation. In this work, we present the first comprehensive overview of the intersection between RL and GUI agents, and examine how this research direction may evolve toward digital inhabitants. We propose a principled taxonomy that organizes existing methods into Offline RL, Online RL, and Hybrid Strategies, and complement it with analyses of reward engineering, data efficiency, and key technical innovations. Our analysis reveals several emerging trends: the tension between reliability and scalability is motivating the adoption of composite, multi-tier reward architectures; GUI I/O latency bottlenecks are accelerating the shift toward world-model-based training, which can yield substantial performance gains; and the spontaneous emergence of System-2-style deliberation suggests that explicit reasoning supervision may not be necessary when sufficiently rich reward signals are available. We distill these findings into a roadmap covering process rewards, continual RL, cognitive architectures, and safe deployment, aiming to guide the next generation of robust GUI automation and its agent-native infrastructure.
| Comments: | Published in TMLR. Project Page: this https URL |
| Subjects: | Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2604.27955 [cs.AI] |
| (or arXiv:2604.27955v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2604.27955 arXiv-issued DOI via DataCite |
|
| Journal reference: | Transactions on Machine Learning Research, 2026 |
Submission history
From: Junan Hu [view email]
[v1]
Thu, 30 Apr 2026 14:51:49 UTC (7,378 KB)
[v2]
Wed, 4 Nov 2026 02:44:36 UTC (7,383 KB)
来源：arXiv cs.AI · arxiv.org