@inproceedings{ding-etal-2025-dynamic,
    title = "Dynamic Parallel Tree Search for Efficient {LLM} Reasoning",
    author = "Ding, Yifu  and
      Jiang, Wentao  and
      Liu, Shunyu  and
      Jing, Yongcheng  and
      Guo, Jinyang  and
      Wang, Yingjie  and
      Zhang, Jing  and
      Wang, Zengmao  and
      Liu, Ziwei  and
      Du, Bo  and
      Liu, Xianglong  and
      Tao, Dacheng",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingestion-acl-25/2025.acl-long.550/",
    pages = "11233--11252",
    ISBN = "979-8-89176-251-0",
    abstract = "Tree of Thoughts (ToT) enhances Large Language Model (LLM) reasoning by structuring problem-solving as a spanning tree. However, recent methods focus on search accuracy while overlooking computational efficiency. The challenges of accelerating the ToT lie in the frequent switching of reasoning focus, and the redundant exploration of suboptimal solutions. To alleviate this dilemma, we propose Dynamic Parallel Tree Search (DPTS), a novel parallelism framework that aims to dynamically optimize the reasoning path in inference. It includes the Parallelism Streamline in the generation phase to build up a flexible and adaptive parallelism with arbitrary paths by cache management and alignment. Meanwhile, the Search and Transition Mechanism filters potential candidates to dynamically maintain the reasoning focus on more possible solutions with less redundancy. Experiments on Qwen-2.5 and Llama-3 on math and code datasets show that DPTS significantly improves efficiency by 2-4$\times$ on average while maintaining or even surpassing existing reasoning algorithms in accuracy, making ToT-based reasoning more scalable and computationally efficient. Codes are released at: https://github.com/yifu-ding/DPTS."
}