重新思考外部慢思考：从雪球误差到正确推理的概率
Rethinking External Slow-Thinking: From Snowball Errors to Probability of Correct Reasoning
- Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China（中国人民大学人工智能学院）
- Beijing Key Laboratory of Research on Large Models（北京大型模型研究关键实验室）
- Engineering Research Center of Next-Generation Intelligent Search（下一代智能搜索工程研究中心）
- College of Artificial Intelligence, Tianjin University of Science（天津科技大学人工智能学院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本文从理论角度研究外部慢思考，揭示其通过降低雪球误差概率提升LLM推理，并指出扩大搜索范围或内部推理能力比框架选择更关键。
AI中文摘要：
测试时扩展（通常也称为慢思考）已被证明能增强大型语言模型（LLMs）中的多步推理能力。然而，尽管其被广泛使用，慢思考方法背后的机制仍知之甚少。本文从理论角度探讨外部慢思考的机制。我们首先考察LLM推理过程中的雪球误差效应，并利用信息论将其与正确推理的可能性联系起来。在此基础上，我们表明外部慢思考方法可以被解释为降低错误概率的策略。我们进一步对流行的外部慢思考方法进行了比较分析，从简单到复杂，突出了它们的差异和相互关系。我们的研究结果表明，这些方法的有效性并非主要取决于所采用的具体框架，而扩大搜索范围或模型的内部推理能力可能在长期内带来更持续的改进。我们在https://github.com/ZyGan1999/Snowball-Errors-and-Probability开源了我们的代码。
英文摘要：
Test-time scaling, which is also often referred to as slow-thinking, has been demonstrated to enhance multi-step reasoning in large language models (LLMs). However, despite its widespread utilization, the mechanisms underlying slow-thinking methods remain poorly understood. This paper explores the mechanisms of external slow-thinking from a theoretical standpoint. We begin by examining the snowball error effect within the LLM reasoning process and connect it to the likelihood of correct reasoning using information theory. Building on this, we show that external slow-thinking methods can be interpreted as strategies to mitigate the error probability. We further provide a comparative analysis of popular external slow-thinking approaches, ranging from simple to complex, highlighting their differences and interrelationships. Our findings suggest that the efficacy of these methods is not primarily determined by the specific framework employed, and that expanding the search scope or the model's internal reasoning capacity may yield more sustained improvements in the long term. We open-source our code at https://github.com/ZyGan1999/Snowball-Errors-and-Probability.
