@inproceedings{leang-etal-2026-picsar,
    title = "{P}i{CSAR}: Probabilistic Confidence Selection and Ranking for Reasoning Chains",
    author = "Leang, Joshua Ong Jun  and
      Zhao, Zheng  and
      Gema, Aryo Pradipta  and
      Yang, Sohee  and
      Kwan, Wai-Chung  and
      He, Xuanli  and
      Li, Wenda  and
      Minervini, Pasquale  and
      Giunchiglia, Eleonora  and
      Cohen, Shay B",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.1577/",
    pages = "31511--31544",
    ISBN = "979-8-89176-395-1",
    abstract = "Best-of-$n$ sampling improves the accuracy of large language models (LLMs) and large reasoning models (LRMs) by generating multiple candidate solutions and selecting the one with the highest reward. The key challenge for reasoning tasks is designing a scoring function that can identify correct reasoning chains without access to ground-truth answers. We propose Probabilistic Confidence Selection and Ranking for Reasoning Chains (PiCSAR): a simple, training-free method that scores each candidate generation using the joint log-likelihood of the reasoning and final answer. This method utilises both the scores of the reasoning path (*reasoning confidence*) and the final answer (*answer confidence*). PiCSAR achieves substantial gains across several benchmarks ($+11.7$ on AIME2024, $+9.81$ on AIME2025), outperforming baselines with at least 2x fewer samples in 20 out of 25 comparisons. Our analysis reveals that correct reasoning chains exhibit higher reasoning and answer confidence, justifying the effectiveness of PiCSAR."
}