让大模型学会在该思考时才思考，避免无意义的过度推理。
Think When You Need: Self-Adaptive Chain-of-Thought Learning
- 基于长度与质量对比构建奖励，动态调节推理深度。
- 在多个基准上保持高准确率，同时解释长度减少40%以上。
- 适用于无标准答案的模糊任务，适合追求效率的部署场景。
链式思维（CoT）推理虽能提升语言模型表现，但常导致简单问题出现低效的“过度思考”。现有方法直接惩罚推理长度，却未考虑问题复杂度差异。本文提出一种通过长度与质量对比构造奖励的新方法，基于理论假设，在保证解题正确性的同时提升简洁性。进一步验证该方法在无标准答案的模糊任务上的有效性。在多个推理基准上的实验表明，该方法在维持准确率的同时，生成的解释显著更短，有效教会模型‘该思考时才思考’。
原文摘要 · Abstract (English)
Chain of Thought (CoT) reasoning enhances language models' performance but often leads to inefficient "overthinking" on simple problems. We identify that existing approaches directly penalizing reasoning length fail to account for varying problem complexity. Our approach constructs rewards through length and quality comparisons, guided by theoretical assumptions that jointly enhance solution correctness with conciseness. Moreover, we further demonstrate our method to fuzzy tasks where ground truth is unavailable. Experiments across multiple reasoning benchmarks demonstrate that our method maintains accuracy while generating significantly more concise explanations, effectively teaching models to "think when needed."
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。