Planner Matters! An Efficient and Unbalanced Multi-agent Collaboration Framework for Long-horizon Planning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911643174174720 |
|---|---|
| author | Wu, Wenyi Zhu, Sibo Zhou, Kun Huang, Biwei |
| author_facet | Wu, Wenyi Zhu, Sibo Zhou, Kun Huang, Biwei |
| contents | Language model (LM)-based agents have demonstrated promising capabilities in automating complex tasks from natural language instructions, yet they continue to struggle with long-horizon planning and reasoning. To address this, we propose an enhanced multi-agent framework that decomposes automation into three roles: a planner for high-level decision-making, an actor for task execution, and a memory manager for contextual reasoning. While this modular decomposition aligns with established design patterns, our core contribution lies in a systematic compute-allocation analysis, revealing that planning is the dominant factor influencing task performance. Execution and memory management require significantly less compute and model capacity to achieve competitive results. Building on these insights, we introduce a planner-centric reinforcement learning approach, which exclusively optimizes the planner using trajectory-level rewards from a VLM-as-judge, while freezing the other components. Extensive experiments on benchmarks spanning web navigation, OS control, and tool use demonstrate that concentrating model capacity and learning on high-level planning yields robust and compute-efficient improvements in long-horizon agent automation. Our code is publicly released. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_02168 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Planner Matters! An Efficient and Unbalanced Multi-agent Collaboration Framework for Long-horizon Planning Wu, Wenyi Zhu, Sibo Zhou, Kun Huang, Biwei Artificial Intelligence Machine Learning Multiagent Systems Language model (LM)-based agents have demonstrated promising capabilities in automating complex tasks from natural language instructions, yet they continue to struggle with long-horizon planning and reasoning. To address this, we propose an enhanced multi-agent framework that decomposes automation into three roles: a planner for high-level decision-making, an actor for task execution, and a memory manager for contextual reasoning. While this modular decomposition aligns with established design patterns, our core contribution lies in a systematic compute-allocation analysis, revealing that planning is the dominant factor influencing task performance. Execution and memory management require significantly less compute and model capacity to achieve competitive results. Building on these insights, we introduce a planner-centric reinforcement learning approach, which exclusively optimizes the planner using trajectory-level rewards from a VLM-as-judge, while freezing the other components. Extensive experiments on benchmarks spanning web navigation, OS control, and tool use demonstrate that concentrating model capacity and learning on high-level planning yields robust and compute-efficient improvements in long-horizon agent automation. Our code is publicly released. |
| title | Planner Matters! An Efficient and Unbalanced Multi-agent Collaboration Framework for Long-horizon Planning |
| topic | Artificial Intelligence Machine Learning Multiagent Systems |
| url | https://arxiv.org/abs/2605.02168 |