Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Han, Qijun, Tu, Haoqin, Wang, Zijun, Dai, Haoyue, Zhou, Yiyang, Lau, Nancy, Cardenas, Alvaro A., Xu, Yuhui, Xu, Ran, Xiong, Caiming, Zheng, Zeyu, Yao, Huaxiu, Zhou, Yuyin, Xie, Cihang
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2604.21375
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914504446574592
author Han, Qijun
Tu, Haoqin
Wang, Zijun
Dai, Haoyue
Zhou, Yiyang
Lau, Nancy
Cardenas, Alvaro A.
Xu, Yuhui
Xu, Ran
Xiong, Caiming
Zheng, Zeyu
Yao, Huaxiu
Zhou, Yuyin
Xie, Cihang
author_facet Han, Qijun
Tu, Haoqin
Wang, Zijun
Dai, Haoyue
Zhou, Yiyang
Lau, Nancy
Cardenas, Alvaro A.
Xu, Yuhui
Xu, Ran
Xiong, Caiming
Zheng, Zeyu
Yao, Huaxiu
Zhou, Yuyin
Xie, Cihang
contents Autonomous GUI agents face two fundamental challenges: early stopping, where agents prematurely declare success without verifiable evidence, and repetitive loops, where agents cycle through the same failing actions without recovery. We present VLAA-GUI, a modular GUI agentic framework built around three integrated components that guide the system on when to Stop, Recover, and Search. First, a mandatory Completeness Verifier enforces UI-observable success criteria and verification at every finish step -- with an agent-level verifier that cross-examines completion claims with decision rules, rejecting those lacking direct visual evidence. Second, a mandatory Loop Breaker provides multi-tier filtering: switching interaction mode after repeated failures, forcing strategy changes after persistent screen-state recurrence, and binding reflection signals to strategy shifts. Third, an on-demand Search Agent searches online for unfamiliar workflows by directly querying a capable LLM with search ability, returning results as plain text. We additionally integrate a Coding Agent for code-intensive actions and a Grounding Agent for precise action grounding, both invoked on demand when required. We evaluate VLAA-GUI across five top-tier backbones, including Opus 4.5, 4.6 and Gemini 3.1 Pro, on two benchmarks with Linux and Windows tasks, achieving top performance on both (77.5% on OSWorld and 61.0% on WindowsAgentArena). Notably, three of the five backbones surpass human performance (72.4%) on OSWorld in a single pass. Ablation studies show that all three proposed components consistently improve a strong backbone, while a weaker backbone benefits more from these tools when the step budget is sufficient. Further analysis also shows that the Loop Breaker nearly halves wasted steps for loop-prone models.
format Preprint
id arxiv_https___arxiv_org_abs_2604_21375
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle VLAA-GUI: Knowing When to Stop, Recover, and Search, A Modular Framework for GUI Automation
Han, Qijun
Tu, Haoqin
Wang, Zijun
Dai, Haoyue
Zhou, Yiyang
Lau, Nancy
Cardenas, Alvaro A.
Xu, Yuhui
Xu, Ran
Xiong, Caiming
Zheng, Zeyu
Yao, Huaxiu
Zhou, Yuyin
Xie, Cihang
Computation and Language
Artificial Intelligence
Software Engineering
Autonomous GUI agents face two fundamental challenges: early stopping, where agents prematurely declare success without verifiable evidence, and repetitive loops, where agents cycle through the same failing actions without recovery. We present VLAA-GUI, a modular GUI agentic framework built around three integrated components that guide the system on when to Stop, Recover, and Search. First, a mandatory Completeness Verifier enforces UI-observable success criteria and verification at every finish step -- with an agent-level verifier that cross-examines completion claims with decision rules, rejecting those lacking direct visual evidence. Second, a mandatory Loop Breaker provides multi-tier filtering: switching interaction mode after repeated failures, forcing strategy changes after persistent screen-state recurrence, and binding reflection signals to strategy shifts. Third, an on-demand Search Agent searches online for unfamiliar workflows by directly querying a capable LLM with search ability, returning results as plain text. We additionally integrate a Coding Agent for code-intensive actions and a Grounding Agent for precise action grounding, both invoked on demand when required. We evaluate VLAA-GUI across five top-tier backbones, including Opus 4.5, 4.6 and Gemini 3.1 Pro, on two benchmarks with Linux and Windows tasks, achieving top performance on both (77.5% on OSWorld and 61.0% on WindowsAgentArena). Notably, three of the five backbones surpass human performance (72.4%) on OSWorld in a single pass. Ablation studies show that all three proposed components consistently improve a strong backbone, while a weaker backbone benefits more from these tools when the step budget is sufficient. Further analysis also shows that the Loop Breaker nearly halves wasted steps for loop-prone models.
title VLAA-GUI: Knowing When to Stop, Recover, and Search, A Modular Framework for GUI Automation
topic Computation and Language
Artificial Intelligence
Software Engineering
url https://arxiv.org/abs/2604.21375