SafeGround: Know When to Trust GUI Grounding Models via Uncertainty Calibration

TL;DR

SafeGround introduces an uncertainty calibration framework for GUI grounding models, enabling risk-aware predictions with controlled false discovery rates, leading to improved system accuracy and reliability.

Contribution

It presents a novel uncertainty-aware calibration method for GUI grounding models that enhances prediction reliability and risk management.

Findings

Outperforms existing baselines in distinguishing correct from incorrect predictions.

Calibrated thresholds enable rigorous risk control and accuracy improvements.

Improves system-level accuracy by up to 5.38 percentage points.

Abstract

Graphical User Interface (GUI) grounding aims to translate natural language instructions into executable screen coordinates, enabling automated GUI interaction. Nevertheless, incorrect grounding can result in costly, hard-to-reverse actions (e.g., erroneous payment approvals), raising concerns about model reliability. In this paper, we introduce SafeGround, an uncertainty-aware framework for GUI grounding models that enables risk-aware predictions through calibrations before testing. SafeGround leverages a distribution-aware uncertainty quantification method to capture the spatial dispersion of stochastic samples from outputs of any given model. Then, through the calibration process, SafeGround derives a test-time decision threshold with statistically guaranteed false discovery rate (FDR) control. We apply SafeGround on multiple GUI grounding models for the challenging ScreenSpot-Pro benchmark. Experimental results show that our uncertainty measure consistently outperforms existing baselines in distinguishing correct from incorrect predictions, while the calibrated threshold reliably enables rigorous risk control and potentials of substantial system-level accuracy improvements. Across multiple GUI grounding models, SafeGround improves system-level accuracy by up to 5.38% percentage points over Gemini-only inference.