PhoneCLI 把 App 图形界面编译成可调用命令，降低移动智能体成本
PhoneCLI: From App Interfaces to Callable Commands for Mobile Agents
这篇论文把 App 导航离线编译成命令，智能体执行又快又省 token，做移动 Agent 的开发者可以直接看看思路。
arXiv 论文提出 PhoneCLI，将 App 的 GUI 导航离线编译为语义化命令，无需应用内部 API、运行时插桩或模型训练。编译后的命令在执行前先验证，然后以亚秒级时间、零 VLM 成本确定性执行。在线阶段智能体选择命令，开放式交互和编译路径失败时回退到内嵌 VLM。在 AndroidLab 基准上 PhoneCLI 提升了任务成功率并减少步骤与 token 消耗，还能迁移到 AndroidWorld 的官方 M3A 智能体并保持效率收益。
PhoneCLI: From App Interfaces to Callable Commands for Mobile Agents
Mobile GUI agents operate through a perception--action loop: at each step they screenshot the device, invoke a vision--language model (VLM), and emit an action. It is slow, costly, and brittle, yet most of what it does is navigation---and everyday navigation is static, ordered, and endlessly repeated. We present PhoneCLI, which compiles an app's GUI navigation into callable commands, without any app-internal API, runtime instrumentation, or model training. Offline, PhoneCLI explores a target app from the outside and distills its screens, interactive elements, and navigation edges into a semantically annotated map; each screen yields one deterministic command: a replay sequence that reaches it. Online, the agent selects a command, verifies it before execution, and then executes it deterministically in sub-second time at zero VLM cost; open-ended interaction and every failure of the compiled path fall back to the embedded VLM interpreter, exactly the pure VLM agent, so compilation can only help. On AndroidLab, PhoneCLI improves the task success rate while reducing steps and token consumption, and it transfers to AndroidWorld's official M3A agent with consistent efficiency gains. What PhoneCLI compiles is the app's navigation rather than one run, so it serves new tasks, not only repeated ones.