Researchers have introduced Qwen-Planner-Agent, a closed-loop AI-for-AI framework designed to develop and iteratively improve real-world mobile planner agents. The work explores whether AI can serve both as the object of development and as an active participant in building next-generation AI systems.
Mobile planning presents a demanding test case: tasks are complex and long-horizon, challenging agent reliability, while costly real-device interaction limits development scalability. To address this, the framework connects data production, model training, and deployment through a shared action-feedback-verification contract.
The framework comprises three components. AI for Data builds a human-gated agentic data flywheel where specialized agents construct tasks, collect interaction trajectories, curate and balance training data, and use training feedback to guide subsequent data generation. AI for Training combines a supervised planning cold start with hybrid-environment online agentic reinforcement learning, introducing Competence-Aware Reward-and-Advantage Engineering (CARE) to reduce reasoning and tool-use costs while preserving task performance. AI drives model-harness co-evolution through an execution-evidence-driven loop that orchestrates memory, skills, and tools at runtime and feeds structured action feedback and preserved failure traces back into coordinated model and harness adaptation.
According to the paper, Qwen-Planner-Agent achieves the best overall performance among all evaluated models and systems on MobilePA-Bench, improving over its base model across tool use, memory, skills, and sub-agent coordination. Further evaluations show improvements across non-mobile agentic benchmarks while largely preserving general capabilities.