PiCSAR: Probabilistic Confidence Selection and Ranking for Reasoning Chains
University of Edinburgh and Imperial College London · University of Edinburgh, University of Edinburgh · University College London, University of London, Department of Computer Science, University College London, University of London, DeepMind and Google · University College London, University of London · University of Edinburgh · Imperial College London
PDF 由论文原始站点提供，PaperCompass 不保存论文文件。DOI 10.18653/v1/2026.findings-acl.1577 ↗
摘要
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.