根据题目难易程度动态调整推理路径数量，大幅减少大模型推理耗时。
Is Discrete Difficulty Sufficient? Leveraging Continuous Difficulty for Efficient Self-Consistency in LLMs
- 用预训练探针预测题目输出熵，连续量化难度并控制生成预算。
- 在多个模型和基准上保持与自洽性相当的准确率，最多节省76% token。
- 适合需要高效推理的场景，如资源受限部署或实时问答系统。
自洽性（Self-Consistency, SC）是一种通过采样多样推理路径并选择最一致答案的解码策略，在复杂推理任务中表现优异。然而，生成多条推理路径带来的高令牌消耗是其主要瓶颈。现有方法虽尝试按问题难度调整路径数或资源分配，但普遍将难度划分为少数固定等级，未能充分捕捉推理复杂度的连续变化特性。本文提出灵活自洽性（Flexible Self-Consistency, FSC），通过预训练探针估计输入问题的输出熵，将其作为模型不确定性的连续指标，动态调节推理路径生成数量。实验表明，在多种模型和基准上，FSC在保持与SC相当准确率的同时，最多可节省76%的令牌消耗。
原文摘要 · Abstract (English)
Self-Consistency (SC) is a decoding strategy that samples diverse reasoning paths and selects the most consistent answer, demonstrating strong performance on complex reasoning problems. However, the excessive token consumption incurred by generating multiple reasoning paths has been identified as a major limitation of SC. To improve computational efficiency, several studies have proposed strategies that adjust the number of reasoning paths or allocate resources differentially according to problem difficulty. Nevertheless, most existing methods categorize difficulty into a few fixed levels, failing to fully capture the continuously varying nature of reasoning complexity. In this work, we propose Flexible Self-Consistency (FSC), which estimates problem difficulty as a continuous signal and dynamically adjusts the number of generated reasoning paths accordingly. FSC predicts the output entropy of an input question using a pre-trained probe and leverages it as an indicator of model uncertainty to flexibly control the sampling budget. Experimental results show that, across various models and benchmarks, FSC maintains accuracy comparable to SC while achieving token savings of up to 76%.
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。