Large Reasoning Models (LRMs) excel at complex reasoning tasks, but their efficiency is often hampered by overly verbose outputs. Prior steering methods attempt to address this issue by applying a single, global vector to hidden representations -- an approach grounded in the restrictive linear representation hypothesis. In this work, we introduce FlowSteer, a nonlinear steering method that goes beyond uniform linear shifts by learning a complete transformation between the distributions associated with verbose and concise reasoning. This transformation is learned via Flow Matching as a velocity field, enabling precise, input-dependent control over the model's reasoning process. By aligning steered representations with the distribution of concise-reasoning activations, FlowSteer yields more compact reasoning than the linear shifts. Across diverse reasoning benchmarks, FlowSteer demonstrates strong task performance and token efficiency compared to leading inference-time baselines. Our work demonstrates that modeling the full distributional transport with generative techniques offers a more effective and principled foundation for controlling LRMs.

翻译：大型推理模型（LRMs）在复杂推理任务中表现出色，但其效率常因输出过于冗长而受限。先前的引导方法试图通过向隐藏表示施加单一的全局向量来解决此问题——这种方法基于限制性的线性表示假设。在本工作中，我们提出了FlowSteer，一种非线性引导方法，它超越了均匀的线性偏移，通过学习与冗长推理和简洁推理相关的分布之间的完整变换来实现。该变换通过流匹配作为速度场进行学习，从而能够对模型的推理过程进行精确的、输入依赖的控制。通过将引导后的表示与简洁推理激活的分布对齐，FlowSteer相比线性偏移能产生更紧凑的推理。在多样化的推理基准测试中，与领先的推理时基线方法相比，FlowSteer展现出强大的任务性能和标记效率。我们的工作表明，利用生成式技术对完整的分布迁移进行建模，为控制LRMs提供了一个更有效且更具原则性的基础。