用命令行比图形界面更高效，能完成更多手机操作任务
Beyond the GUI Paradigm: Do Mobile Agents Need the Phone Screen?
- 直接调用手机命令行接口，绕过图形界面限制
- 命令行代理在45项日常任务中平均用10.7步完成，远少于图形界面的18.6步
- 适合研究移动智能体新范式或想提升自动化效率的开发者
当前移动智能体多依赖图形用户界面（GUI）感知与交互，但安卓平台也提供可直接访问系统服务与数据的命令行接口（CLI）。本文主张将CLI作为与GUI同等重要的范式。在AndroidWorld和MobileWorld两个数据集上，未经过移动端特定微调的三个编码代理（Claude Code、Terminus-2、mini-swe-agent）对比多个模型API，结果显示Claude Code（Opus 4.7）在AndroidWorld达71.8%、MobileWorld达51.9%，超越所有可复现的GUI基线（最高69.3%/68.1%/57.8%；43.2%/26.3%/13.3%）。通过构造理想化指令解法，最高可达88.8%（103/116任务可解）和86.3%（101/117任务可解），显示仍有巨大提升空间。为覆盖GUI无法覆盖的用户需求，提出包含45个模板的「CLI优势任务套件」，涵盖批量操作、多条件筛选、聚合、跨应用流程及隐藏设备状态五类场景。所有CLI代理在五类任务中均优于所有GUI基线，且任务步数显著减少（10.7 vs. 18.6）。研究代码、真值解法、任务套件与评估框架将开源。
原文摘要 · Abstract (English)
Recent advances in mobile agents are dominated by the GUI paradigm, in which agents perceive UI information and emit screen interactions. However, mobile platforms also expose a command-line interface (CLI) that provides direct access to device services and data. We argue CLI deserves first-class consideration alongside GUI. We evaluate three coding agents (Claude Code, Terminus-2, mini-swe-agent) across four model APIs on AndroidWorld and MobileWorld without any mobile-specific post-training, comparing against three reproducible GUI baselines (GUI-Owl-1.5-32B, MAI-UI, Qwen3-VL-32B). Claude Code (Opus 4.7) reaches 71.8\% and 51.9\%, outperforming every reproducible GUI baseline (69.3/68.1/57.8\% on AndroidWorld; 43.2/26.3/13.3\% on MobileWorld), while every other CLI configuration remains competitive. To establish the paradigm's ceiling, we provide oracle CLI solutions that reach 88.8\% on AndroidWorld (103/116 tasks CLI-solvable) and 86.3\% on MobileWorld (101/117 tasks CLI-solvable), indicating substantial room for future improvement. To cover everyday user intents beyond the GUI scope, we introduce the \textbf{CLI-Advantage Task Suite}, comprising 45 templates across five categories: bulk operations, multi-condition filtering, aggregation, cross-app workflows, and hidden device state. Every CLI agent outperforms every GUI baseline in all five categories, with substantially fewer steps per task (10.7 vs.\ 18.6). To support future research on mobile CLI agents, we will open-source agent implementations, oracle solutions, the CLI-Advantage suite, and evaluation infrastructure.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。