Orcust: Stepwise-Feedback Reinforcement Learning for GUI Agent

📅 2025-09-22
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
Existing GUI agents face two critical challenges: unreliable reward signals and limited capacity for online trajectory generation—leading to unreliable reasoning and low data efficiency. To address these, we propose Orcust, a novel framework introducing two key mechanisms: (1) Principle-Constrained Reward Modeling (PCRM), which incorporates interpretable, human-aligned rules into the reinforcement learning reward function to ensure policy adherence to domain principles; and (2) Online Virtual Machine-Grounded Trajectory Construction (OVTC), which leverages a lightweight virtual machine to autonomously generate high-quality, structured interaction trajectories enabling fine-grained, stepwise training. Integrating chain-of-thought reasoning with LLM-driven rule feedback, Orcust achieves +22.2% and +23.9% improvements on ScreenSpot and ScreenSpot-Pro benchmarks, respectively. The framework significantly enhances reasoning reliability, task adaptability, and system scalability.

Technology Category

Application Category

📝 Abstract
Recent advances in GUI agents have achieved remarkable grounding and action-prediction performance, yet existing models struggle with unreliable reward signals and limited online trajectory generation. In this paper, we introduce Orcust, a framework that integrates Principle-Constrained Reward Modeling (PCRM) and Online VM-Grounded Trajectory Construction (OVTC) to enhance reasoning reliability and data efficiency in interactive GUI tasks. We leverages environment-verifiable and LLM-derived principle to enforce interpretable reward signals that constrain long chain-of-thought reasoning and rule-based feedback. OVTC spins up instrumented virtual machines to autonomously collect structured GUI interaction trajectories with explicit procedural and structural objectives, enabling the training of a stepwise reward model that robustly captures human preferences and adheres to task-specific constraints. Extensive experiments on standard GUI benchmarks covering perceptual grounding, foundational operations, and end-to-end task execution reveal that Orcust achieves state-of-the-art performance, improving by 22.2% on ScreenSpot and 23.9% on ScreenSpot-Pro over the base model (i.e. Qwen2.5-VL-7B). The results demonstrate Orcust's effectiveness in enhancing the reasoning, adaptability and scalability of GUI agents across various environments and task complexities.
Problem

Research questions and friction points this paper is trying to address.

GUI agents struggle with unreliable reward signals during training
Limited online trajectory generation hinders interactive GUI performance
Existing models lack robust reasoning reliability and data efficiency
Innovation

Methods, ideas, or system contributions that make the work stand out.

Principle-Constrained Reward Modeling for interpretable signals
Online VM-Grounded Trajectory Construction for data collection
Stepwise reward model capturing human preferences and constraints
J
Junyu Lu
Lionrock AI Lab, China Merchants Research Institute of Advanced Technology
S
Songxin Zhang
Lionrock AI Lab, China Merchants Research Institute of Advanced Technology
Z
Zejian Xie
Lionrock AI Lab, China Merchants Research Institute of Advanced Technology
Z
Zhuoyang Song
Lionrock AI Lab, China Merchants Research Institute of Advanced Technology
J
Jiaxing Zhang
Lionrock AI Lab, China Merchants Research Institute of Advanced Technology