Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.11170 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916652585582592 |
|---|---|
| author | Xu, Yibin Yang, Liang Chen, Hao Wang, Hua Chen, Zhi Tang, Yaohua |
| author_facet | Xu, Yibin Yang, Liang Chen, Hao Wang, Hua Chen, Zhi Tang, Yaohua |
| contents | The limitation of graphical user interface (GUI) data has been a significant barrier to the development of GUI agents today, especially for the desktop / computer use scenarios. To address this, we propose an automated GUI data generation pipeline, AutoCaptioner, which generates data with rich descriptions while minimizing human effort. Using AutoCaptioner, we created a novel large-scale desktop GUI dataset, DeskVision, along with the largest desktop test benchmark, DeskVision-Eval, which reflects daily usage and covers diverse systems and UI elements, each with rich descriptions. With DeskVision, we train a new GUI understanding model, GUIExplorer. Results show that GUIExplorer achieves state-of-the-art (SOTA) performance in understanding/grounding visual elements without the need for complex architectural designs. We further validated the effectiveness of the DeskVision dataset through ablation studies on various large visual language models (LVLMs). We believe that AutoCaptioner and DeskVision will significantly advance the development of GUI agents, and will open-source them for the community. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_11170 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DeskVision: Large Scale Desktop Region Captioning for Advanced GUI Agents Xu, Yibin Yang, Liang Chen, Hao Wang, Hua Chen, Zhi Tang, Yaohua Computation and Language The limitation of graphical user interface (GUI) data has been a significant barrier to the development of GUI agents today, especially for the desktop / computer use scenarios. To address this, we propose an automated GUI data generation pipeline, AutoCaptioner, which generates data with rich descriptions while minimizing human effort. Using AutoCaptioner, we created a novel large-scale desktop GUI dataset, DeskVision, along with the largest desktop test benchmark, DeskVision-Eval, which reflects daily usage and covers diverse systems and UI elements, each with rich descriptions. With DeskVision, we train a new GUI understanding model, GUIExplorer. Results show that GUIExplorer achieves state-of-the-art (SOTA) performance in understanding/grounding visual elements without the need for complex architectural designs. We further validated the effectiveness of the DeskVision dataset through ablation studies on various large visual language models (LVLMs). We believe that AutoCaptioner and DeskVision will significantly advance the development of GUI agents, and will open-source them for the community. |
| title | DeskVision: Large Scale Desktop Region Captioning for Advanced GUI Agents |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2503.11170 |