Computer-Use Agent
Papers
- Wang, Xinyuan, et al. “Opencua: Open foundations for computer-use agents.” NeurIPS, 2026. [paper] [code] [dataset] [tool] [demo]
- Liu, Zhaoyang, et al. “Scalecua: Scaling open-source computer use agents with cross-platform data.” arXiv preprint arXiv:2509.15221 (2025). [paper] [code] [dataset] [demo]
- Qin, Yujia, et al. “Ui-tars: Pioneering automated gui interaction with native agents.” arXiv preprint arXiv:2501.12326 (2025). [pdf] [code]
- EvoCUA: Evolving Computer Use Agents via Learning from Scalable Synthetic Experience [technical report] [code]
- Xu, Haiyang, et al. “Mobile-agent-v3. 5: Multi-platform fundamental gui agents.” arXiv preprint arXiv:2602.16855 (2026). [paper] [code]
- Zhang, Bofei, et al. “Tongui: Internet-scale trajectories from multimodal web tutorials for generalized gui agents.” AAAI, 2026. [pdf] [code] [dataset]
Datasets
- OSWorld: evaluation, online
- AgentNet: training and evaluation, manual annotation
- ScaleCUA-Data: training, synthetic
- GUI-Net-1M: training, created from web tutorials