Metraipolis › archive › dc_40249d490a66
Qwen-Planner-Agent: A Closed-Loop AI-for-AI Framework for Real-World Mobile Planner Agents — arXiv:2609.29892
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.29892
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T04:26:25.179Z
- SHA-256 of the archived text
e7180bcf1ac6be6a2cf5ee50011ef61c1e50a6f0b16148b2d33c9f268bae0811- Stable URL of this page
https://metraipolis.com/documento/dc_40249d490a66
How to cite
Qwen-Planner-Agent: A Closed-Loop AI-for-AI Framework for Real-World Mobile Planner Agents — arXiv:2609.29892 — arXiv (en), https://arxiv.org/abs/2609.29892 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_40249d490a66 (Metraipolis archive dc_40249d490a66, sha256 e7180bcf1ac6be6a2cf5ee50011ef61c1e50a6f0b16148b2d33c9f268bae0811)
BibTeX
@misc{dc40249d490a66,
title = {Qwen-Planner-Agent: A Closed-Loop AI-for-AI Framework for Real-World Mobile Planner Agents — arXiv:2609.29892},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.29892},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_40249d490a66; sha256 e7180bcf1ac6be6a2cf5ee50011ef61c1e50a6f0b16148b2d33c9f268bae0811},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.29892. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
The rapid progression of large language models is extending AI from passive content generation into the active workflows of engineering and scientific discovery. This shift raises a compelling question: can AI be both the object of development and an active participant in building next-generation AI systems? We explore this question by building Qwen-Planner-Agent within a closed-loop AI-for-AI framework for scalable development and iterative improvement. Mobile planning offers a demanding test of this approach: complex, long-horizon tasks challenge agent reliability, while costly real-device interaction limits development scalability. The framework connects data production, model training, and deployment through a shared action-feedback-verification contract. (i) AI for Data builds a human-gated agentic data flywheel in which specialized agents construct tasks, collect interaction trajectories, curate and balance training data, and use training feedback to guide subsequent data generation. (ii) AI for Training combines a supervised planning cold start with hybrid-environment online agentic reinforcement learning, where we introduce Competence-Aware Reward-and-Advantage Engineering (CARE) to reduce reasoning and tool-use costs while preserving task performance. (iii) 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. 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 of our model show improvements across non-mobile agentic benchmarks while largely preserving general capabilities. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-25 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.29892 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-08T03:50:57.849Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).