Steering Large Reasoning Models towards Concise Reasoning via Flow Matching

2026-02-05

AI Summary

Although large reasoning models excel at complex tasks, their lengthy outputs often lead to inefficiency. This work proposes FlowSteer, a novel approach that, for the first time, models the transformation of reasoning processes as a velocity field, thereby moving beyond conventional linear representation assumptions. By leveraging flow matching techniques, FlowSteer enables nonlinear and dynamic guidance of hidden state distributions. The method supports input-dependent, fine-grained control and demonstrates significant improvements in both task performance and token efficiency across multiple reasoning benchmarks, outperforming existing inference-time steering methods.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

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.

Problem

• Large Reasoning Models
• concise reasoning
• verbose outputs
• reasoning efficiency
• output verbosity

Innovation

• FlowSteer
• Flow Matching
• nonlinear steering
• Large Reasoning Models
• distributional transport