arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.05539cs.LGcs.AIcs.CL
收藏

通过流匹配引导大型推理模型实现紧凑推理

Steering Large Reasoning Models towards Concise Reasoning via Flow Matching

LMU Munich(慕尼黑莱茵河大学);Amazon(亚马逊)

Yawei Li, Benjamin Bergner, Yinghan Zhao, Vihang Prakash Patil, Bei Chen, Cheng Wang

AI总结:

通过流匹配引导大型推理模型实现紧凑推理,提出非线性引导方法提升推理效率和任务性能。

摘要:

大型推理模型(LRMs)在复杂推理任务中表现出色,但其效率常受到过于冗长输出的限制。先前的引导方法试图通过应用一个全局向量到隐藏表示来解决这一问题,这种方法基于线性表示假设的限制。在本工作中,我们引入了FlowSteer,一种非线性引导方法,它超越了均匀线性位移,通过学习verbose和concise推理相关分布之间的完整转换来实现。该转换通过流匹配作为速度场进行学习,使模型的推理过程能够实现精确的输入依赖控制。通过将引导表示与紧凑推理激活分布对齐,FlowSteer实现了比线性位移更紧凑的推理。在多样化的推理基准测试中,FlowSteer在任务性能和令牌效率方面均优于领先的推理时间基线。我们的工作表明,使用生成技术建模完整的分布传输能够为控制LRMs提供更有效和原则性的基础。

英文摘要:

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.

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。

↑