相同任务，不同应用：移动GUI智能体为何难以泛化？
Same Tasks, Different Apps: Why Mobile GUI Agents Fail to Generalize?
- KAIST（韩国科学技术院）
- CMU（卡内基梅隆大学）
机构由 AI 辅助整理，请以论文原文为准。
中文总结 AI 辅助
提出AnyAppBench基准，在保持用户目标不变下评估移动GUI智能体跨应用泛化能力，发现13个智能体在原始应用上的成功无法可靠迁移到新应用，高层子目标分解仅带来微小变化，为稳健部署提供垫脚石。
中文摘要 AI 辅助
在实际环境中部署的移动GUI智能体必须能够在支持相同功能的不同应用程序中工作。现有的大多数基准测试仅在单一应用中测试每个任务，因此高分可能意味着智能体理解了任务，或者仅仅意味着它熟悉那个特定的应用。我们引入了AnyAppBench，一个类别控制的实时安卓基准测试，在保持用户目标固定的同时评估跨应用的泛化能力。该基准涵盖10个功能类别、100个任务模板和52个应用中的520个任务-应用对。智能体从原始指令运行，并带有与应用无关的子目标，一个VLM评判员在固定的失败分类体系下对每次失败运行进行标注，该分类体系的可靠性通过人工标注来衡量。我们发现，在13个智能体中，在原始应用上的成功并不能可靠地迁移到具有相同目标的新应用上。此外，提供高层子目标分解仅产生微小的、依赖类别的变化，无法弥合差距，且失败类型的混合随目标界面而变化。基于这些见解，我们认为AnyAppBench基准为移动GUI智能体的稳健实际部署提供了重要的垫脚石。我们的代码、数据和排行榜可在项目网站上找到，网址为https://this https URL。
英文摘要
Mobile GUI agents deployed in real settings must work across different applications that support the same functionality. Most existing benchmarks test each task in only one app, so a high score can mean the agent understands the task, or only that it knows that particular app. We introduce AnyAppBench, a category-controlled live Android benchmark that evaluates cross-application generalization while keeping the user goal fixed. It spans 10 functional categories, 100 task templates, and 520 task--application pairs over 52 applications. Agents run from raw instructions and with app-independent sub-goals, and a VLM judge labels every failed run under a fixed failure taxonomy whose reliability is measured by human annotation. We find that, across 13 agents, success on the original application does not transfer reliably to new applications with the same goal. Furthermore, providing high-level sub-goal decomposition produces only small, category-dependent changes that do not close the gap, and the mix of failure types changes with the target interface. Based on those insights, we believe the AnyAppBench benchmark provides an important stepping stone toward robust real-world deployment of mobile GUI agents. Our code, data and the leaderboard can be found at the project website https://anyappbench.github.io/.
