MobileRL: Online Agentic Reinforcement Learning for Mobile GUI Agents
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914112148078592 |
|---|---|
| author | Xu, Yifan Liu, Xiao Liu, Xinghan Fu, Jiaqi Zhang, Hanchen Jing, Bohao Zhang, Shudan Wang, Yuting Zhao, Wenyi Dong, Yuxiao |
| author_facet | Xu, Yifan Liu, Xiao Liu, Xinghan Fu, Jiaqi Zhang, Hanchen Jing, Bohao Zhang, Shudan Wang, Yuting Zhao, Wenyi Dong, Yuxiao |
| contents | Building general-purpose graphical user interface (GUI) agents has become increasingly promising with the progress in vision language models. However, developing effective mobile GUI agents with reinforcement learning (RL) remains challenging due to the heavy-tailed distribution of task difficulty and the inefficiency of large-scale environment sampling. We present an online agentic reinforcement learning framework MobileRL to enhance GUI agents in mobile environments. Its core component is the Difficulty-ADAptive GRPO (ADAGRPO) algorithm. In ADAGRPO, we design difficulty-adaptive positive replay and failure curriculum filtering to adapt the model to different task difficulties. We introduce the shortest-path reward adjustment strategy to reshape rewards concerning the task length in multi-turn agentic tasks. Those strategies jointly stabilize RL training, improve sample efficiency, and generate strong performance across diverse mobile apps and tasks. We apply MOBILERL to two open models (Qwen2.5-VL-7B-Instruct and GLM-4.1V-9B-Base). The resultant MOBILERL-9B model achieves state-of-the-art results in terms of success rates on both AndroidWorld (80.2%) and AndroidLab (53.6%). The MOBILERL framework is open-sourced at: https://github.com/THUDM/MobileRL. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_18119 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MobileRL: Online Agentic Reinforcement Learning for Mobile GUI Agents Xu, Yifan Liu, Xiao Liu, Xinghan Fu, Jiaqi Zhang, Hanchen Jing, Bohao Zhang, Shudan Wang, Yuting Zhao, Wenyi Dong, Yuxiao Machine Learning Artificial Intelligence Building general-purpose graphical user interface (GUI) agents has become increasingly promising with the progress in vision language models. However, developing effective mobile GUI agents with reinforcement learning (RL) remains challenging due to the heavy-tailed distribution of task difficulty and the inefficiency of large-scale environment sampling. We present an online agentic reinforcement learning framework MobileRL to enhance GUI agents in mobile environments. Its core component is the Difficulty-ADAptive GRPO (ADAGRPO) algorithm. In ADAGRPO, we design difficulty-adaptive positive replay and failure curriculum filtering to adapt the model to different task difficulties. We introduce the shortest-path reward adjustment strategy to reshape rewards concerning the task length in multi-turn agentic tasks. Those strategies jointly stabilize RL training, improve sample efficiency, and generate strong performance across diverse mobile apps and tasks. We apply MOBILERL to two open models (Qwen2.5-VL-7B-Instruct and GLM-4.1V-9B-Base). The resultant MOBILERL-9B model achieves state-of-the-art results in terms of success rates on both AndroidWorld (80.2%) and AndroidLab (53.6%). The MOBILERL framework is open-sourced at: https://github.com/THUDM/MobileRL. |
| title | MobileRL: Online Agentic Reinforcement Learning for Mobile GUI Agents |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2509.18119 |