[2511.22235] Training High-Level Schedulers with Execution-Feedback Reinforcement Learning for Long-Horizon GUI Automation

Computer Science > Artificial Intelligence arXiv:2511.22235 (cs) [Submitted on 27 Nov 2025 (v1), last revised 4 Mar 2026 (this version, v2)] Title:Training High-Level Schedulers with Execution-Feedback Reinforcement Learning for Long-Horizon GUI Automation Authors:Zehao Deng, Tianjie Ju, Zheng Wu, Zhuosheng Zhang, Gongshen Liu View a PDF of the paper titled Training High-Level Schedulers with Execution-Feedback Reinforcement Learning for Long-Horizon GUI Automation, by Zehao Deng and 4 other authors View PDF HTML (experimental) Abstract:The rapid development of large vision-language model (VLM) has greatly promoted the research of GUI agent. However, GUI agents still face significant challenges in handling long-horizon tasks. First, single-agent models struggle to balance high-level capabilities and low-level execution capability, facing prevalent issues of responsibility coupling and capability conflicts. Second, agents lack awareness of the task state, leading to progress loss in long-horizon tasks. To address these challenges, we propose a staged execution-feedback reinforcement learning algorithm. Unlike training a unified policy model, we focus on training high-level scheduling models. Specifically, we propose and train two agents: a Coordinator, responsible for the strategic planning and task decomposition; and a State Tracker, responsible for context compression and information management to maintain the task's state and coherence. Based on this, we built the Coordi...