{"ID":6537726,"CreatedAt":"2026-07-14T02:54:43.516908796Z","UpdatedAt":"2026-07-15T03:28:55.185153975Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.11185","arxiv_id":"2607.11185","title":"SCALECUA: Scaling Computer Use Agents with Verifiable Task Synthesis and Efficient Online RL","abstract":"Computer use agents (CUAs) are emerging as a powerful interface for automating complex digital workflows through visual perception and GUI execution. Online reinforcement learning with verifiable rewards (RLVR) has emerged as a key direction for scaling their capabilities. However, this paradigm is bottlenecked by verifiable data scarcity and online RL inefficiency. To break these barriers, we introduce ScaleCUA, a unified framework that scales online RL for CUAs via verifiable task synthesis and efficient training. At the data level, we design VeriGen, an end-to-end framework for generating verifiable RL tasks through iterative docker interactions and a multi-agent feedback loop. Scaled to 100+ concurrent agent workers via a shared docker interaction probe, this pipeline produces 24K+ verifiable tasks and nearly 3K high-quality RL tasks. To maximize sample efficiency, we propose Frontier Sampling, which tracks per-task capability and allocates rollouts to the current learning frontier. On the training side, we further design Visual Context Segmentation, a sliding window over recent visual context that balances rollout and training-engine pressure, yielding a 2.83x training speedup over step-wise decomposition. Together, ScaleCUA achieves 68.7% on OSWorld and 54.0% on ScienceBoard, establishing new state-of-the-art performance among open-source computer use agents. Code, models, and datasets are available at https://github.com/THUDM/SCALE-CUA.","short_abstract":"Computer use agents (CUAs) are emerging as a powerful interface for automating complex digital workflows through visual perception and GUI execution. Online reinforcement learning with verifiable rewards (RLVR) has emerged as a key direction for scaling their capabilities. However, this paradigm is bottlenecked by veri...","url_abs":"https://arxiv.org/abs/2607.11185","url_pdf":"https://arxiv.org/pdf/2607.11185v1","authors":"[\"Bowen Lv\",\"Xiao Liu\",\"Yanyu Ren\",\"Hanyu Lai\",\"Bohao Jing\",\"Hanchen Zhang\",\"Yanxiao Zhao\",\"Shuntian Yao\",\"Jie Tang\",\"Yuxiao Dong\"]","published":"2026-07-13T07:32:35Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Reinforcement Learning\"]","has_code":false,"code_links":[{"ID":614226,"CreatedAt":"2026-07-14T02:54:43.516908796Z","UpdatedAt":"2026-07-14T02:54:43.516908796Z","DeletedAt":null,"paper_id":6537726,"paper_url":"https://arxiv.org/abs/2607.11185","paper_title":"SCALECUA: Scaling Computer Use Agents with Verifiable Task Synthesis and Efficient Online RL","repo_url":"https://github.com/THUDM/SCALE-CUA","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}