用熵值动态调整大模型推理深度，提升复杂问题求解效率。
Entropy-based Exploration Conduction for Multi-step Reasoning
- 通过监控输出熵和熵方差动态控制推理深度
- 在四个基准数据集上均提升推理准确率与效率
- 适合需要精细探索的复杂逻辑推理任务
大语言模型在多步推理中表现优异，但推理深度对任务性能影响显著。现有自动决策方法成本高且缺乏灵活性。为此，我们提出基于熵的探索深度引导方法（Entro-duction），通过监测模型输出熵及熵的方差来捕捉当前步骤的不确定性及其在连续步骤间的波动性。基于熵的变化，模型以概率方式选择深化、扩展或终止探索，实现推理准确率与探索效率的平衡。在四个基准数据集上的实验表明，该方法有效提升了多步推理性能。
原文摘要 · Abstract (English)
Multi-step processes via large language models (LLMs) have proven effective for solving complex reasoning tasks. However, the depth of exploration of the reasoning procedure can significantly affect the task performance. Existing methods to automatically decide the depth often lead to high cost and a lack of flexibility. To address these issues, we propose Entropy-based Exploration Depth Conduction (Entro-duction), a novel method that dynamically adjusts the exploration depth during multi-step reasoning by monitoring LLM's output entropy and variance entropy. We employ these two features to capture the model's uncertainty of the current step and the fluctuation of uncertainty across consecutive reasoning steps. Based on the observed entropy changes, the LLM selects whether to deepen, expand, or stop exploration according to the probability, which facilitates the trade-off between the reasoning accuracy and exploration effectiveness. Experimental results across four benchmark datasets demonstrate the efficacy of Entro-duction.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。