ScreenAgent: A Vision Language Model-driven Computer Control Agent
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915304348581888 |
|---|---|
| author | Niu, Runliang Li, Jindong Wang, Shiqi Fu, Yali Hu, Xiyu Leng, Xueyuan Kong, He Chang, Yi Wang, Qi |
| author_facet | Niu, Runliang Li, Jindong Wang, Shiqi Fu, Yali Hu, Xiyu Leng, Xueyuan Kong, He Chang, Yi Wang, Qi |
| contents | Existing Large Language Models (LLM) can invoke a variety of tools and APIs to complete complex tasks. The computer, as the most powerful and universal tool, could potentially be controlled directly by a trained LLM agent. Powered by the computer, we can hopefully build a more generalized agent to assist humans in various daily digital works. In this paper, we construct an environment for a Vision Language Model (VLM) agent to interact with a real computer screen. Within this environment, the agent can observe screenshots and manipulate the Graphics User Interface (GUI) by outputting mouse and keyboard actions. We also design an automated control pipeline that includes planning, acting, and reflecting phases, guiding the agent to continuously interact with the environment and complete multi-step tasks. Additionally, we construct the ScreenAgent Dataset, which collects screenshots and action sequences when completing a variety of daily computer tasks. Finally, we trained a model, ScreenAgent, which achieved computer control capabilities comparable to GPT-4V and demonstrated more precise UI positioning capabilities. Our attempts could inspire further research on building a generalist LLM agent. The code is available at \url{https://github.com/niuzaisheng/ScreenAgent}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_07945 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ScreenAgent: A Vision Language Model-driven Computer Control Agent Niu, Runliang Li, Jindong Wang, Shiqi Fu, Yali Hu, Xiyu Leng, Xueyuan Kong, He Chang, Yi Wang, Qi Human-Computer Interaction Artificial Intelligence Computer Vision and Pattern Recognition Existing Large Language Models (LLM) can invoke a variety of tools and APIs to complete complex tasks. The computer, as the most powerful and universal tool, could potentially be controlled directly by a trained LLM agent. Powered by the computer, we can hopefully build a more generalized agent to assist humans in various daily digital works. In this paper, we construct an environment for a Vision Language Model (VLM) agent to interact with a real computer screen. Within this environment, the agent can observe screenshots and manipulate the Graphics User Interface (GUI) by outputting mouse and keyboard actions. We also design an automated control pipeline that includes planning, acting, and reflecting phases, guiding the agent to continuously interact with the environment and complete multi-step tasks. Additionally, we construct the ScreenAgent Dataset, which collects screenshots and action sequences when completing a variety of daily computer tasks. Finally, we trained a model, ScreenAgent, which achieved computer control capabilities comparable to GPT-4V and demonstrated more precise UI positioning capabilities. Our attempts could inspire further research on building a generalist LLM agent. The code is available at \url{https://github.com/niuzaisheng/ScreenAgent}. |
| title | ScreenAgent: A Vision Language Model-driven Computer Control Agent |
| topic | Human-Computer Interaction Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2402.07945 |