Jean-Pierre Lo

Do Multi-Agents Dream of Electric Screens? Achieving Perfect Accuracy on AndroidWorld Through Task Decomposition

We present Minitap, a multi-agent system that achieves 100% success on the AndroidWorld benchmark, the first to fully solve all 116 tasks and surpassing human performance (80%). We first analyze why single-agent architectures fail: context pollution from mixed reasoning traces, silent text input failures undetected by the agent, and repetitive action loops without escape. Minitap addresses each failure through targeted mechanisms: cognitive separation across six specialized agents, deterministic post-validation of text input against device state, and meta-cognitive reasoning that detects cycles and triggers strategy changes. Ablations show multi-agent decomposition contributes +21 points over single-agent baselines; verified execution adds +7 points; meta-cognition adds +9 points. We release Minitap as open-source software. https://github.com/minitap-ai/mobile-use