الملخص
تمت ترجمة أجزاء من هذه الصفحة آلياً وقد تحتوي على أخطاء.
Computer-use agents (CUAs) have advanced along two separate lines: graphical interaction and software development through code and the command line. Real digital work requires both, interleaved rather than stacked end to end. We study hybrid CUAs that autonomously decide when to explore an interface, implement software, and run and visually verify their artifacts. We introduce RecreationWorld, a five-platform framework built around recreation: given a running reference, an agent must discover its behavior and build a faithful implementation with no prescribed workflow. RecreationWorld provides reproducible environments on Ubuntu, macOS, Windows, Android, and Web, plus a unified harness with native GUI control and coding tools. The running reference serves as an oracle for hidden behavioral tests, providing execution-grounded rewards. We scale trajectory generation with high-quality open-source applications. Models trained on these trajectories improve across five out-of-distribution coding and hybrid computer-use benchmarks and more frequently verify their rendered outputs, providing evidence of transfer beyond recreation. For held-out evaluation, we introduce RecreationBench, comprising 250 diverse tasks across domains and platforms. Reference-grounded programmatic and visual assertions cover action-conditioned outcomes at multiple interaction depths; each is validated on the reference and by human reviewers before the suite is frozen for automatic scoring. GPT-6 Astra leads at 58.1% overall, but passes all programmatic tests on just 2.8% of tasks. Agents reproduce static interface structure more reliably than interactions and computed outputs, while generated applications remain smaller and more monolithic than their references. We release the benchmark, environments, and test suites.
الكلمات المفتاحية
الموضوع
بيانات النشر
- المجلة
- غير متاح
- وصول مفتوح
- وصول مفتوح أخضر
اقتبس هذه المقالة
APA 7
Bai, S., Deng, J., Fan, S., Fu, Y., Gao, C., Hu, X., Huang, M., Jiang, Y., Jing, Y., Kong, D., Li, K., Li, N., Li, W., Liu, D., Lu, D., Luo, C., Shen, Q., Wang, Z., Wang, Z., Wu, J., Wu, G., Xie, Z., Xie, R., Xu, H., Yang, A., Yuan, J., Zhang, Y., Zhang, J., Zhang, X., Zhang, Z., Zhen, Z., Zhu, M., & Zhou, B. (2026). RecreationWorld: بيئات قابلة للتوسع والتحقق لوكلاء استخدام الحاسوب الهجينة. https://omanscience.com/ar/articles/recreationworld-scalable-and-verifiable-environments-for-hybrid-computer-use-agents
MLA 9
Bai, Shuai, et al. "RecreationWorld: بيئات قابلة للتوسع والتحقق لوكلاء استخدام الحاسوب الهجينة." https://omanscience.com/ar/articles/recreationworld-scalable-and-verifiable-environments-for-hybrid-computer-use-agents.
شيكاغو (المؤلف–التاريخ)
Bai, Shuai, Jiayong Deng, Sicheng Fan, Yikun Fu, Chang Gao, Xuhao Hu, Mianqiu Huang, Yizhen Jiang, Yuheng Jing, Dehui Kong, Keliang Li, Ning Li, Wanli Li, Dayiheng Liu, Dunjie Lu, Changwei Luo, Que Shen, Zheyuan Wang, Zijian Wang, Jie Wu, Gao Wu, Zhihui Xie, Rui Xie, Haiyang Xu, An Yang, Jiakang Yuan, Yanming Zhang, Jiajun Zhang, Xi Zhang, Zhenru Zhang, Zhuo Zhen, Mingkang Zhu, and Bowen Zhou. 2026. "RecreationWorld: بيئات قابلة للتوسع والتحقق لوكلاء استخدام الحاسوب الهجينة." https://omanscience.com/ar/articles/recreationworld-scalable-and-verifiable-environments-for-hybrid-computer-use-agents.
هارفارد
Bai, S., Deng, J., Fan, S., Fu, Y., Gao, C., Hu, X., Huang, M., Jiang, Y., Jing, Y., Kong, D., Li, K., Li, N., Li, W., Liu, D., Lu, D., Luo, C., Shen, Q., Wang, Z., Wang, Z., Wu, J., Wu, G., Xie, Z., Xie, R., Xu, H., Yang, A., Yuan, J., Zhang, Y., Zhang, J., Zhang, X., Zhang, Z., Zhen, Z., Zhu, M. and Zhou, B. (2026) 'RecreationWorld: بيئات قابلة للتوسع والتحقق لوكلاء استخدام الحاسوب الهجينة', Available at: https://omanscience.com/ar/articles/recreationworld-scalable-and-verifiable-environments-for-hybrid-computer-use-agents.
فانكوفر
Bai S, Deng J, Fan S, Fu Y, Gao C, Hu X, et al. RecreationWorld: بيئات قابلة للتوسع والتحقق لوكلاء استخدام الحاسوب الهجينة. https://omanscience.com/ar/articles/recreationworld-scalable-and-verifiable-environments-for-hybrid-computer-use-agents
IEEE
S. Bai, J. Deng, S. Fan, Y. Fu, C. Gao, X. Hu, M. Huang, Y. Jiang, Y. Jing, D. Kong, K. Li, N. Li, W. Li, D. Liu, D. Lu, C. Luo, Q. Shen, Z. Wang, Z. Wang, J. Wu, G. Wu, Z. Xie, R. Xie, H. Xu, A. Yang, J. Yuan, Y. Zhang, J. Zhang, X. Zhang, Z. Zhang, Z. Zhen, M. Zhu, and B. Zhou, "RecreationWorld: بيئات قابلة للتوسع والتحقق لوكلاء استخدام الحاسوب الهجينة," https://omanscience.com/ar/articles/recreationworld-scalable-and-verifiable-environments-for-hybrid-computer-use-agents.