paper-with-me

홈 › Papers

Temporal UI State Inconsistency in Desktop GUI Agents: Formalizing and Defending Against TOCTOU Attacks on Computer-Use Agents

2026-04-20 · Wenpeng Xu arxiv

GUI agents that control desktop computers via screenshot-and-click loops introduce a new class of vulnerability: the observation-to-action gap (mean 6.51 s on real OSWorld workloads) creates a Time-Of-Check, Time-Of-Use (TOCTOU) window during which an unprivileged attacker can manipulate the UI state. We formalize this as a Visual Atomicity Violation and characterize three concrete attack primitives: (A) Notification Overlay Hijack, (B) Window Focus Manipulation, and (C) Web DOM Injection. Primitive B, the closest desktop analog to Android Action Rebinding, achieves 100% action-redirection success rate with zero visual evidence at the observation time. We propose Pre-execution UI State Verification (PUSV), a lightweight three-layer defense that re-verifies the UI state immediately before each action dispatch: masked pixel SSIM at the click target (L1), global screenshot diff (L2a), and X Window snapshot diff (L2b). PUSV achieves 100% Action Interception Rate across 180 adversarial trials (135 Primitive A + 45 Primitive B) with zero false positives and < 0.1 s overhead. Against Primitive C (zero-visual-footprint DOM injection), PUSV reveals a structural blind spot (~0% AIR), motivating future OS+DOM defense-in-depth architectures. No single PUSV layer alone achieves full coverage; different primitives require different detection signals, validating the layered design.

📄 PDF Abstract BibTeX arXiv:2604.18860

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

UI-Vision: A Desktop-centric GUI Benchmark for Visual Perception and Interaction

2025-03-19 · Shravan Nayak, Xiangru Jian, Kevin Qinghong Lin, Juan A. Rodriguez 외

Autonomous agents that navigate Graphical User Interfaces (GUIs) to automate tasks like document editing and file management can greatly enhance computer workflows. While existing research focuses on online settings, des…

NavigateSpatial Reasoning

ComputerRL: Scaling End-to-End Online Reinforcement Learning for Computer Use Agents

2025-08-19 · Hanyu Lai, Xiao Liu, Yanxiao Zhao, Han Xu 외 arxiv

We introduce ComputerRL, a framework for autonomous desktop intelligence that enables agents to operate complex digital workspaces skillfully. ComputerRL features the API-GUI paradigm, which unifies programmatic API call…

Reinforcement Learning

DeskVision: Large Scale Desktop Region Captioning for Advanced GUI Agents

2025-03-14 · Yibin Xu, Liang Yang, Hao Chen, Hua Wang 외

The limitation of graphical user interface (GUI) data has been a significant barrier to the development of GUI agents today, especially for the desktop / computer use scenarios. To address this, we propose an automated G…

ChainWorld: Composing Long-Horizon Desktop Workloads from Atomic OSWorld Tasks

2026-06-19 · Vincent Siu, Manasi Sharma, Dawn Song, Daniel Yue Zhang 외 arxiv

Computer use agents are evaluated almost exclusively on atomic desktop tasks, but realistic desktop work requires sustaining state across multiple objectives. We study this gap with ChainWorld, which composes atomic OSWo…

ANCHOR: Branch-Point Data Generation for GUI Agents

2026-02-06 · Jinbiao Wei, Yilun Zhao, Kangqi Ni, Arman Cohan arxiv

End-to-end GUI agents for real desktop environments require large amounts of high-quality interaction data, yet collecting human demonstrations is expensive and existing synthetic pipelines often suffer from limited task…