Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI Agents

TL;DR

This paper introduces UGround, a universal visual grounding model trained on a large GUI dataset, enabling GUI agents to perceive and interact with interfaces visually, outperforming existing models and agents.

Contribution

The paper presents a novel visual grounding approach for GUI agents, including a large dataset, a simple training recipe, and a model that surpasses existing methods in multiple benchmarks.

Findings

UGround outperforms existing visual grounding models by up to 20%.

Agents with UGround outperform state-of-the-art agents using only visual input.

The approach demonstrates the feasibility of human-like visual navigation in GUI agents.

Abstract

Multimodal large language models (MLLMs) are transforming the capabilities of graphical user interface (GUI) agents, facilitating their transition from controlled simulations to complex, real-world applications across various platforms. However, the effectiveness of these agents hinges on the robustness of their grounding capability. Current GUI agents predominantly utilize text-based representations such as HTML or accessibility trees, which, despite their utility, often introduce noise, incompleteness, and increased computational overhead. In this paper, we advocate a human-like embodiment for GUI agents that perceive the environment entirely visually and directly perform pixel-level operations on the GUI. The key is visual grounding models that can accurately map diverse referring expressions of GUI elements to their coordinates on the GUI across different platforms. We show that a simple recipe, which includes web-based synthetic data and slight adaptation of the LLaVA architecture, is surprisingly effective for training such visual grounding models. We collect the largest dataset for GUI visual grounding so far, containing 10M GUI elements and their referring expressions over 1.3M screenshots, and use it to train UGround, a strong universal visual grounding model for GUI agents. Empirical results on six benchmarks spanning three categories (grounding, offline agent, and online agent) show that 1) UGround substantially outperforms existing visual grounding models for GUI agents, by up to 20% absolute, and 2) agents with UGround outperform state-of-the-art agents, despite the fact that existing agents use additional text-based input while ours only uses visual perception. These results provide strong support for the feasibility and promises of GUI agents that navigate the digital world as humans do.