AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

MobileRL: Online Agentic Reinforcement Learning for Mobile GUI Agents

Yifan Xu, Xiao Liu, Xinghan Liu, Jiaqi Fu, Jiayu Huang, Hanchen Zhang, Bohao Jing, Shudan Zhang, Yuting Wang, Zhao wenyi, Yuxiao Dong

ICLR 2026 · Reinforcement learning › Visual and representation RL · approach: Scalable RL systems and environments · 53 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add AdaGRPO RL stage after SFT (MobileRL-9B) (ablation)MobileRL-9B · AndroidWorldsuccess rate (%): 66.2 -> 80.2Improved
add AdaGRPO RL stage after SFT (MobileRL-9B) (ablation)MobileRL-9B · AndroidLabsuccess rate (%): 45.0 -> 53.6Improved
add RL stage (ablation)MobileRL-9B · AndroidControl-HighSR (%): 77.5 -> 73.9Hurt
add RL stage (ablation)MobileRL-7B · AndroidControl-LowSR (%): 91.9 -> 91.4No clear effect
add RL stage (ablation)MobileRL-7B · AndroidControl-Highgrounding: 73.6 -> 71.5Hurt
add Reasoning-Free SFT before Reasoning SFT (ablation)MobileRL · AndroidWorldsuccess rate (%)Improved
add AdaPR replay (ablation)MobileRL · AndroidWorldreward / success rateImproved
add SPA reward adjustment (ablation)MobileRL · AndroidWorldstep efficiencyImproved
add FCF failure curriculum filtering (ablation)MobileRL · AndroidWorldrewardImproved
use original instead of compressed images (GLM-4.1V-9B) (ablation)MobileRL-9B AdaGRPO · AndroidWorldsuccess rate (%): 75.8 -> 80.2Improved
use original instead of compressed images (Qwen2.5-VL) (ablation)MobileRL-7B · AndroidWorldsuccess rate (%)No clear effect
use original instead of compressed images (ablation)Reasoning SFT GLM-4.1V-9B · AndroidWorldsuccess rate (%): 66.1 -> 66.2No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy