Steering vectors offer a training-free mechanism for controlling reasoning behaviors in large language models, but constructing effective vectors requires identifying genuine behavioral signals in the model's hidden states. For behaviors that can be toggled via prompts, this is straightforward. However, many reasoning behaviors -- such as self-reflection -- emerge spontaneously and resist prompt-level control. Current methods detect these behaviors through keyword matching in chain-of-thought traces, implicitly assuming that every detected boundary encodes a genuine behavioral signal. We show that this assumption is overwhelmingly wrong: across 541 keyword-detected boundaries, 93.3\% are behaviorally unstable, failing to reproduce the detected behavior under re-generation from the same prefix. We develop a probabilistic model that formalizes intrinsic reasoning behaviors as stochastic events with context-dependent trigger probabilities, and show that unstable boundaries dilute the steering signal. Guided by this analysis, we propose stability filtering, which retains only boundaries where the model consistently reproduces the target behavior. Combined with a content-subspace projection that removes residual question-specific noise, our method achieves 0.784 accuracy on MATH-500 (+5.0 over the strongest baseline). The resulting steering vectors transfer across models in the same architecture family without re-extraction, improving Nemotron-Research-Reasoning-1.5B (+5.0) and DeepScaleR-1.5B-Preview (+6.0). Code is available at https://github.com/zhmzm/stability-steering.

翻译：引导向量为大语言模型中控制推理行为提供了免训练机制，但构建有效向量需要从模型隐藏状态中识别真实的行为信号。对于可通过提示词切换的行为，该过程较为直接。然而，诸如自我反思等许多推理行为会自发涌现，难以通过提示词进行控制。现有方法通过链式思维轨迹中的关键词匹配检测这些行为，并隐含假设每个被检测到的边界都编码了真实的行为信号。我们证明该假设存在重大偏差：在541个关键词检测边界中，93.3%的行为不稳定，在相同前缀下的重新生成过程中无法复现已检测到的行为。我们提出一个概率模型，将内在推理行为形式化为具有上下文相关触发概率的随机事件，并证明不稳定边界会削弱引导信号。基于该分析，我们提出稳定性过滤方法，仅保留模型能一致复现目标行为的边界。结合用于消除剩余问题特定噪声的内容子空间投影，我们的方法在MATH-500上达到0.784的准确率（较最强基线提升5.0）。最终得到的引导向量无需重新提取即可在同一架构系列的模型间迁移，使Nemotron-Research-Reasoning-1.5B提升5.0，DeepScaleR-1.5B-Preview提升6.0。代码开源地址：https://github.com/zhmzm/stability-steering