HyperTree Planning: Enhancing LLM Reasoning via Hierarchical Thinking

Runquan Gui1, Zhihai Wang1,2, Jie Wang1, Chi Ma1, Huiling Zhen2, Mingxuan Yuan2, Jianye Hao2, Defu Lian1, Enhong Chen1, Feng Wu1
1 University of Science and Technology of China;
2 Noah’s Ark Lab, Huawei Technologies;
International Conference on Machine Learning (ICML) 2025
arXiv Code

Abstract

Recent advancements have significantly enhanced the performance of large language models (LLMs) in tackling complex reasoning tasks, achieving notable success in domains like mathematical and logical reasoning. However, these methods encounter challenges with complex planning tasks, primarily due to extended reasoning steps, diverse constraints, and the challenge of handling multiple distinct sub-tasks. To address these challenges, we propose HyperTree Planning (HTP), a novel reasoning paradigm that constructs hypertree-structured planning outlines for effective planning. The hypertree structure enables LLMs to engage in hierarchical thinking by flexibly employing the divide-and-conquer strategy, effectively breaking down intricate reasoning steps, accommodating diverse constraints, and managing multiple distinct sub-tasks in a well-organized manner. We further introduce an autonomous planning framework that completes the planning process by iteratively refining and expanding the hypertree-structured planning outlines. Experiments demonstrate the effectiveness of HTP, achieving state-of-the-art accuracy on the TravelPlanner benchmark with Gemini-1.5-Pro, resulting in a 3.6× performance improvement over o1-preview.

HTP: From Tree Planning to HyperTree Planning

Experiments on Complex Planning Tasks

To evaluate the effectiveness of our method, we select three of the most challenging planning datasets: Travel Planner, PlanBench and Natural Plan.

BibTeX

@inproceedings{gui2025hypertree,
  title={HyperTree Planning: Enhancing LLM Reasoning via Hierarchical Thinking},
  author={Gui, Runquan and Wang, Zhihai and Wang, Jie and Ma, Chi and Zhen, Huiling and Yuan, Mingxuan and Hao, Jianye and Lian, Defu and Chen, Enhong and Wu, Feng},
  booktitle={Proceedings of the 42nd International Conference on Machine Learning},
  year={2025}
}