DRS-GUI: Dynamic Region Search for Training-Free GUI Grounding

TL;DR

DRS-GUI is a training-free framework that improves GUI element grounding by dynamically exploring interface regions using human-like perceptual actions and Monte Carlo Tree Search, enhancing performance of multimodal language models.

Contribution

It introduces a novel, training-free dynamic region search method with a perceptual action planner for better GUI grounding in multimodal models.

Findings

Achieves 14% improvement on ScreenSpot-Pro benchmarks.

Significantly enhances grounding performance and generalization.

Effectively integrates into existing MLLMs without additional training.

Abstract

GUI agents powered by Multimodal Large Language Models (MLLMs) have demonstrated impressive capability in understanding and executing user instructions. However, accurately grounding instruction-relevant elements from high-resolution screenshots cluttered with irrelevant UI components remains challenging for existing approaches. Inspired by how humans dynamically adjust their perceptual scope to locate task-related regions on complex screens, we propose DRS-GUI, a training-free dynamic region search framework for GUI grounding that can be seamlessly integrated into existing MLLMs. DRS-GUI introduces a lightweight UI Perceptor that performs three human-like perceptual actions (Focus, Shift, and Scatter) to progressively explore the interface and generate region proposals. To dynamically schedule these actions, we further design an Action Planner based on Monte Carlo Tree Search (MCTS). A region quality reward is employed to evaluate and select the highly instruction-relevant region, efficiently pruning redundant UI elements. Experiments demonstrate that DRS-GUI yields a 14\% improvement on ScreenSpot-Pro for general and GUI-specific MLLMs (Qwen2.5-VL-7B and UGround-V1-7B), significantly enhancing grounding performance and generalization.