重新审视有限监督下的思维链推理：半监督思维链学习
Revisiting Chain-of-Thought Reasoning under Limited Supervision: Semi-supervised Chain-of-Thought Learning
AI总结：
提出半监督思维链学习框架Semi-CoT，利用未标注问题通过熵门控选择低熵伪思维链作为可靠监督信号，实验显示高精度但部分任务效果有限。
摘要：
思维链推理已成为激活大型语言模型潜在推理能力的有效方法。然而，现有大多数CoT方法主要将推理链用作推理时的提示，而生成的推理轨迹很少被重用为半监督学习信号。在本报告中，我们定义了\textbf{半监督思维链学习}并提出了\textbf{Semi-CoT}，一个简单的框架，使用未标注问题构建伪推理监督。Semi-CoT为每个未标注问题采样多个伪CoT，估计答案级语义熵，并选择低熵推理链作为可靠的伪CoT演示。这将CoT的自我训练视角从推理时优化扩展到半监督伪监督。在AQuA、SVAMP、GSM8K和MultiArith上的初步实验表明，熵门控选择了高精度的伪CoT，伪答案精度范围从$91.36\%$到$100\%$。Semi-CoT在SVAMP和GSM8K上也有小幅提升，而AQuA出现负迁移，MultiArith达到上限。这些结果表明，未标注问题可以提供可靠的伪推理信号，但其有效使用仍需更强的演示选择或学生训练。
英文摘要：
Chain-of-thought (CoT) reasoning has emerged as an effective approach for activating latent reasoning capabilities in large language models. However, most existing CoT methods use reasoning chains mainly as inference-time prompts, while the generated reasoning traces are rarely reused as semi-supervised learning signals. In this report, we define \textbf{Semi-supervised Chain-of-Thought Learning} and propose \textbf{Semi-CoT}, a simple framework that uses unlabeled questions to construct pseudo reasoning supervision. Semi-CoT samples multiple pseudo-CoTs for each unlabeled question, estimates answer-level semantic entropy, and selects low-entropy reasoning chains as reliable pseudo-CoT demonstrations. This extends the self-training view of CoT from inference-time refinement to semi-supervised pseudo-supervision. Pilot experiments on AQuA, SVAMP, GSM8K, and MultiArith show that the entropy gate selects high-precision pseudo-CoTs, with pseudo-answer precision ranging from $91.36\%$ to $100\%$. Semi-CoT also gives small gains on SVAMP and GSM8K, while AQuA shows negative transfer and MultiArith reaches a ceiling. These results suggest that unlabeled questions can provide reliable pseudo reasoning signals, but their effective use still requires stronger demonstration selection or student training.
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。
