Large language models (LLMs) equipped with chain-of-thought (CoT) achieve strong performance and offer a window into LLM behavior. However, recent evidence suggests that improvements in CoT capabilities often come with redundant reasoning processes, motivating a key question: Can LLMs acquire a fast-thinking mode analogous to human System 1 reasoning? To explore this, our study presents a self-sampling framework based on activation steering for efficient CoT learning. Our method can induce style-aligned and variable-length reasoning traces from target LLMs themselves without any teacher guidance, thereby alleviating a central bottleneck of SFT-based methods-the scarcity of high-quality supervision data. Using filtered data by gold answers, we perform SFT for efficient CoT learning with (i) a human-like dual-cognitive system, and (ii) a progressive compression curriculum. Furthermore, we explore a self-evolution regime in which SFT is driven solely by prediction-consistent data of variable-length variants, eliminating the need for gold answers. Extensive experiments on math benchmarks, together with cross-domain generalization tests in medicine, show that our method yields stable improvements for both general and R1-style LLMs. Our data and model checkpoints can be found at https://github.com/DYR1/S3-CoT.

翻译：配备思维链（CoT）的大语言模型（LLM）展现出强劲性能，并为理解LLM行为提供了窗口。然而，近期研究表明，CoT能力的提升常伴随冗余的推理过程，这引出一个关键问题：LLM能否获得类似人类系统1快速思维的推理模式？为探索此问题，本研究提出一种基于激活引导的自采样框架，用于高效CoT学习。该方法能够从目标LLM自身诱导出风格对齐且长度可变的推理轨迹，无需任何教师指导，从而缓解了基于监督微调（SFT）方法的核心瓶颈——高质量监督数据的稀缺性。利用通过标准答案筛选的数据，我们通过以下两种机制执行SFT以实现高效CoT学习：（i）类人双认知系统；（ii）渐进式压缩课程。此外，我们探索了一种自进化机制，其中SFT完全由预测一致的可变长度变体数据驱动，无需依赖标准答案。在数学基准测试上的大量实验，连同医学领域的跨领域泛化测试均表明，我们的方法能为通用型及R1风格LLM带来稳定提升。相关数据与模型检查点可在 https://github.com/DYR1/S3-CoT 获取。