Jev-Mobile: Jev as an Executor for Mobile GUI Agents· Jev-Mobile: 移动 GUI 代理的执行器
Vision-language models (VLMs) have become a common foundation for autonomous mobile GUI agents, but most existing systems rely on the VLM for both planning and action grounding at nearly every interaction step, leading to substantial latency and model-serving cost. We introduce Jev-Mobile, which shifts this paradigm to low-frequency VLM planning and high-frequency lightweight execution: the VLM specifies local goals, the accessibility tree defines a structured executable action space, and Jev, a fast typed decision model, repeatedly selects actions within this space. This design allows multiple GUI actions to be executed under a single VLM decision, reducing expensive VLM inference while preserving adaptive interaction. On the full AndroidWorld task suite, Jev-Mobile achieves 79% task succ
使用 VLM 规划、轻量级模型执行，提高移动 GUI 代理交互性能。
- 核心方法
- 引入 Jev-Mobile，采用低频 VLM 规划与高频轻量级执行结合，VLM 制定局部目标，Jev 在可访问性树定义的动作空间内选取动作。
- 适合谁读
- 研究者 / 工程师 / 产品
- 要解决的问题
- 现有的移动 GUI 代理系统依赖 VLM 进行规划和执行，导致高延迟和模型服务成本。
- 关键实验
- 在 AndroidWorld 任务套件上进行了测试，结果显示 Jev-Mobile 达到 79% 的任务成功率。
- 主要贡献
- 显著降低 VLM 推理成本，同时保持交互的适应性，任务成功率高达 79%。
- 意义与局限
- 提高了移动 GUI 代理的效率和可用性，但可能在某些复杂任务上表现有限。