熵轨迹形状预测LLM推理可靠性：对不确定性动态的诊断研究
Entropy trajectory shape predicts LLM reasoning reliability: A diagnostic study of uncertainty dynamics in chain-of-thought
AI总结：
研究通过分析推理过程中熵轨迹形状来预测大语言模型的推理可靠性，提出了一种基于轨迹形状而非标量大小的诊断方法，展示了其在黑盒环境中的实用性与鲁棒性。
摘要：
理解链式推理中的不确定性对于可靠部署大型语言模型至关重要。在本工作中，我们提出了一种基于轨迹形状而非标量大小的简单而有效的诊断方法。我们证明了该信号在黑盒环境中实用、可解释且成本低廉，同时在不同模型和数据集上保持鲁棒性。通过广泛的消融研究和跨领域复现，我们展示了其在选择性预测和优先级排序中的用途。我们的发现为推理任务中的不确定性动态提供了可推广的见解，特别关注数值和离散答案设置。
英文摘要：
Understanding uncertainty in chain-of-thought reasoning is critical for reliable deployment of large language models. In this work, we propose a simple yet effective diagnostic approach based on trajectory shape rather than scalar magnitude. We show that this signal is practical, interpretable, and inexpensive to obtain in black-box settings, while remaining robust across models and datasets. Through extensive ablations and cross-domain replications, we demonstrate its utility for selective prediction and triage. Our findings offer a generalizable insight into uncertainty dynamics in reasoning tasks, with particular focus on numeric and discrete-answer settings.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
