@inproceedings{liu-etal-2026-equivpruner,
    title = "{E}quiv{P}runer: Boosting Efficiency and Quality in {LLM}-Based Search via Action Pruning",
    author = "Liu, Jiawei  and
      Chen, Qisi  and
      Zhang, Jianshu  and
      Liu, Quan  and
      Lian, Defu",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.100/",
    pages = "2211--2226",
    ISBN = "979-8-89176-390-6",
    abstract = "Large Language Models (LLMs) excel at complex reasoning through search algorithms, yet current strategies often suffer from massive token consumption due to redundant exploration of semantically equivalent steps. Existing semantic similarity methods struggle to accurately identify such equivalence in domain-specific contexts like mathematical reasoning. To address this, we propose \textit{EquivPruner}, a simple yet effective approach that identifies and prunes semantically equivalent actions during LLM reasoning search. We also introduce MathEquiv, the first dataset we created for mathematical statement equivalence, which enables the training of a lightweight equivalence detector. Extensive experiments across various models and tasks demonstrate that \textit{EquivPruner} significantly reduces token consumption, improving searching efficiency and often bolstering reasoning accuracy. For instance, when applied to Qwen2.5-Math-7B-Instruct on GSM8K, \textit{EquivPruner} reduced token consumption by 48.1{\%} while also improving accuracy. Our code is available at \url{https://github.com/Lolo1222/EquivPruner}."
}