@inproceedings{nie-etal-2026-attnpo,
    title = "{A}ttn{PO}: Attention-Guided Process Supervision for Efficient Reasoning",
    author = "Nie, Shuaiyi  and
      Dingsiyu  and
      Zhang, Wenyuan  and
      Yu, Linhao  and
      Yang, Tianmeng  and
      Chen, Yao  and
      Yin, Weichong  and
      Sun, Yu  and
      Wu, Hua  and
      Liu, Tingwen",
    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.1845/",
    pages = "39728--39748",
    ISBN = "979-8-89176-390-6",
    abstract = "Large reasoning models trained with reinforcement learning and verifiable rewards (RLVR) achieve strong performance on complex reasoning tasks, yet often overthink, generating redundant reasoning without performance gains. Existing trajectory-level length penalties often fail to effectively shorten reasoning length and degrade accuracy, as they uniformly treat all reasoning steps and lack fine-grained signals to distinguish redundancy from necessity. Meanwhile, process-supervised methods are typically resource-intensive and suffer from inaccurate credit assignment. To address these issues, we propose ATTNPO, a low-overhead process-supervised RL framework that leverages the model{'}s intrinsic attention signals for step-level credit assignment. We first identify a set of special attention heads that naturally focus on essential steps while suppressing redundant ones. By leveraging the attention scores of these heads, We then employ two sub-strategies to mitigate overthinking by discouraging redundant steps while preserving accuracy by reducing penalties on essential steps. Experimental results show that ATTNPO substantially reduces reasoning length while significantly improving performance across 9 benchmarks."
}