@inproceedings{huang-etal-2026-sat,
    title = "{SAT}: Balancing Reasoning Accuracy and Efficiency with Stepwise Adaptive Thinking",
    author = "Huang, Weiyang  and
      Bai, Xuefeng  and
      Chen, Kehai  and
      Chen, Xinyang  and
      Chen, Yibin  and
      Guan, Weili  and
      Zhang, Min",
    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.2009/",
    pages = "43384--43402",
    ISBN = "979-8-89176-390-6",
    abstract = "Large Reasoning Models (LRMs) have revolutionized complex problem-solving, yet they exhibit a pervasive ``overthinking'', generating unnecessarily long reasoning chains. While current solutions improve token efficiency, they often sacrifice fine-grained control or risk disrupting the logical integrity of the reasoning process. To address this, we introduce Stepwise Adaptive Thinking (SAT), a framework that performs step-level, difficulty-aware pruning while preserving the core reasoning structure. SAT formulates reasoning as a Finite-State Machine (FSM) with distinct thinking modes (SLOW, NORMAL, FAST, SKIP). It navigates these states dynamically using a lightweight Process Reward Model (PRM), compressing easy steps while preserving depth for hard ones. Experiments across 9 LRMs and 7 benchmarks show that SAT achieves up to 40{\%} reduction in reasoning tokens while generally maintaining or improving accuracy. Code is available at https://github.com/byxw13/SAT{\_}Code."
}