Don't Act Blindly: Robust GUI Automation via Action-Effect Verification and Self-Correction

Autonomous GUI Agents based on visionlanguage models (VLMs) often assume deterministic environment responses, generating actions without verifying whether previous operations succeeded. In real-world settings with network latency, rendering delays, and system interruptions, this assumption leads to undetected action failures, repetitive ineffective behaviors, and catastrophic error accumulation. Moreover, learning robust recovery strategies is challenging due to the high cost of online interaction and the lack of real-time feedback in offline datasets. We propose VeriGUI (Verification-driven GUI Agent), which explicitly models action outcomes and recovery under noisy environments. VeriGUI introduces a Thinking-Verification-Action-Expectation (TVAE) framework to detect failures and guide corrective reasoning, and a two-stage training pipeline that combines Robust SFT with synthetic failure trajectories and GRPO with asymmetric verification rewards. We train VeriGUI in both 3B and 7B configurations. We further construct a Robustness Benchmark based on AndroidControl-High to evaluate failure recognition and correction. Experiments demonstrate that VeriGUI-3B and -7B achieve strong results across offline and online benchmarks, with the verification-andrecovery mechanism transferring effectively to dynamic real-world environments.