Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI Agents
By Boyu Gou, Ruohan Wang, Boyuan Zheng, Yanan Xie, Cheng Chang, Yiheng Shu, Huan Sun, Yu Su. International Conference on Learning Representations, 2024
374 citations recorded by the source corpus.
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.
Subfields: Artificial Intelligence, Social Psychology
View Paper: https://www.semanticscholar.org/paper/a15a4cabe7653a87efc3302552fbc88b25d19dc1
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- Zhiyong Wu
- Zhenyu Wu
- Fangzhi Xu
- Yian Wang
- Chengyou Jia
- Qiushi Sun
- Zichen Ding
- Liheng Chen
Citations in the corpus
- Group-in-Group Policy Optimization for LLM Agent Training (2025, neutral): “For instance, they now power embodied assistants that navigate simulated homes [5, 6], mobile and web navigators that plan multi-step browsing sessions [7, 8, 9], and autonomous explorers in interactive games [10, 11].”
- OS-ATLAS: A Foundation Action Model for Generalist GUI Agents (2024, methodology): “A concurrent study by Gou et al. (2024) also addresses these challenges; however, their focus is limited to scaling web data.”
- Leveraging Vision-Language Models for Visual Grounding and Analysis of Automotive UI (2025, unknown): “To determine if our fine-tuning resulted in overfitting to the automotive UI domain, we applied ELAM-7B to the ScreenSpot dataset [3] comparing it to SeeClick [3], ShowUI [18], OS-Atlas-Base-7B [33], Molmo-7B-D-0924/Molmo-72B-0924 [4], and Qwen2-VL-based UGround-V1-7B [10] in Table 4.”