English

CoT-Space: A Theoretical Framework for Internal Slow-Thinking via Reinforcement Learning

Artificial Intelligence 2025-09-26 v2 Computation and Language

Abstract

Reinforcement Learning (RL) has become a pivotal approach for enhancing the reasoning capabilities of Large Language Models (LLMs). However, a significant theoretical gap persists, as traditional token-level RL frameworks fail to align with the reasoning-level nature of complex, multi-step thought processes like Chain-of-Thought (CoT). To address this challenge, we introduce CoT-Space, a novel theoretical framework that recasts LLM reasoning from a discrete token-prediction task to an optimization process within a continuous, reasoning-level semantic space. This shift in perspective serves as a conceptual bridge, revitalizing foundational principles from classical learning theory to analyze the unique dynamics of LLMs. By analyzing this process from both a noise perspective and a risk perspective, we demonstrate that the convergence to an optimal CoT length is a natural consequence of the fundamental trade-off between underfitting and overfitting. Furthermore, extensive experiments provide strong empirical validation for our theoretical findings. Our framework not only provides a coherent explanation for empirical phenomena such as overthinking but also offers a solid theoretical foundation to guide the future development of more effective and generalizable reasoning agents. We open-source our code at https://github.com/ZyGan1999/CoT-Space.

Keywords

Cite

@article{arxiv.2509.04027,
  title  = {CoT-Space: A Theoretical Framework for Internal Slow-Thinking via Reinforcement Learning},
  author = {Zeyu Gan and Hao Yi and Yong Liu},
  journal= {arXiv preprint arXiv:2509.04027},
  year   = {2025}
}

Comments

Preprint Edition

R2 v1 2026-07-01T05:20:44.066Z