ScaleCUA: Scaling Open-Source Computer Use Agents with Cross-Platform Data
taesiri
Abstract
ScaleCUA, a large-scale dataset and model for computer use agents, achieves state-of-the-art performance across multiple platforms and tasks by leveraging data-driven scaling.
Vision-Language Models (VLMs) have enabled computer use agents (CUAs) that operate GUIs autonomously, showing great potential, yet progress is limited by the lack of large-scale, open-source computer use data and foundation models. In this work, we introduce ScaleCUA, a step toward scaling open-source CUAs. It offers a large-scale dataset spanning 6 operating systems and 3 task domains, built via a closed-loop pipeline uniting automated agents with human experts. Trained on this scaled-up data, ScaleCUA can operate seamlessly across platforms. Specifically, it delivers strong gains over baselines (+26.6 on WebArena-Lite-v2, +10.7 on ScreenSpot-Pro) and sets new state-of-the-art results (94.4% on MMBench-GUI L1-Hard, 60.6% on OSWorld-G, 47.4% on WebArena-Lite-v2). These findings underscore the power of data-driven scaling for general-purpose computer use agents. We will release data, models, and code to advance future research: https://github.com/OpenGVLab/ScaleCUA.
Community
Vision-Language Models (VLMs) have enabled computer use agents (CUAs) that operate GUIs autonomously, showing great potential, yet progress is limited by the lack of large-scale, open-source computer use data and foundation models. In this work, we introduce ScaleCUA, a step toward scaling open-source CUAs. It offers a large-scale dataset spanning 6 operating systems and 3 task domains, built via a closed-loop pipeline uniting automated agents with human experts. Trained on this scaled-up data, ScaleCUA can operate seamlessly across platforms. Specifically, it delivers strong gains over baselines (+26.6 on WebArena-Lite-v2, +10.7 on ScreenSpot-Pro) and sets new state-of-the-art results (94.4% on MMBench-GUI L1-Hard, 60.6% on OSWorld-G, 47.4% on WebArena-Lite-v2). These findings underscore the power of data-driven scaling for general-purpose computer use agents.
For a deeper understanding, check out the arXiv page for this paper: https://arxiv.org/abs/1905.11946
arXiv explained breakdown of this paper 👉 https://arxivexplained.com/papers/scalecua-scaling-open-source-computer-use-agents-with-cross-platform-data
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- OpenCUA: Open Foundations for Computer-Use Agents (2025)
- VeriOS: Query-Driven Proactive Human-Agent-GUI Interaction for Trustworthy OS Agents (2025)
- ComputerRL: Scaling End-to-End Online Reinforcement Learning for Computer Use Agents (2025)
- Towards Adaptive ML Benchmarks: Web-Agent-Driven Construction, Domain Expansion, and Metric Optimization (2025)
- OS-MAP: How Far Can Computer-Using Agents Go in Breadth and Depth? (2025)
- Chain-of-Agents: End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RL (2025)
- WebSight: A Vision-First Architecture for Robust Web Agents (2025)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Here's the arXiv explained breakdown of this paper: https://arxivexplained.com/papers/scalecua-scaling-open-source-computer-use-agents-with-cross-platform-data
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper