用熵值动态截断推理步骤，让小模型更省力又准
EntroCut: Entropy-Guided Adaptive Truncation for Efficient Chain-of-Thought Reasoning in Small-scale Large Reasoning Models
- 根据早期推理的输出熵判断是否可提前结束
- 最多节省40%计算量，准确率损失极小
- 适合追求效率的轻量化大模型推理场景
大型推理模型（LRMs）通过生成长链式思维在复杂任务上表现优异，但冗长的中间步骤带来巨大计算开销。我们发现，模型早期推理步骤中输出分布的熵能可靠区分正确与错误推理路径。基于此，提出无需训练的EntroCut方法，通过识别高置信度状态动态截断推理过程，安全终止。为全面评估效率与准确率的权衡，引入统一指标效率-性能比（EPR），量化每单位准确率损失对应的令牌节省量。在四个基准测试中，EntroCut将令牌使用量减少高达40%，且相比现有无训练方法实现了更优的效率-性能平衡。结果表明，基于熵的动态截断是缓解LRM低效问题的实用方案。
原文摘要 · Abstract (English)
Large Reasoning Models (LRMs) excel at complex reasoning tasks through extended chain-of-thought generation, but their reliance on lengthy intermediate steps incurs substantial computational cost. We find that the entropy of the model's output distribution in early reasoning steps reliably distinguishes correct from incorrect reasoning. Motivated by this observation, we propose EntroCut, a training-free method that dynamically truncates reasoning by identifying high-confidence states where reasoning can be safely terminated. To comprehensively evaluate the trade-off between efficiency and accuracy, we introduce the Efficiency-Performance Ratio (EPR), a unified metric that quantifies relative token savings per unit accuracy loss. Experiments on four benchmarks show that EntroCut reduces token usage by up to 40\% with minimal accuracy sacrifice, achieving superior efficiency-performance trade-offs compared with existing training-free methods. These results demonstrate that entropy-guided dynamic truncation provides a practical approach to mitigate the inefficiency of LRMs.
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。