提出推理模型的计算与准确率规律，提升逻辑一致性
When Reasoning Meets Its Laws
- 构建推理规律框架，要求计算量随问题复杂度线性增长
- 实测多数模型满足单调性但缺乏组合性，性能受限
- 通过微调强化组合性，显著提升多任务推理表现
尽管大型推理模型（LRMs）表现优异，其推理行为常显反直觉，导致推理能力不足。本文提出推理规律（LoRe）框架，统一刻画LRM内在推理模式。首先提出计算律：推理计算量应随问题复杂度线性增长；进一步引入准确率律。由于问题复杂度难以量化，通过单调性和组合性两个可测性质验证假设。为此构建LoRe-Bench基准，系统评估大模型的这两项性质。实验显示，多数模型具备合理单调性但缺乏组合性。针对此，提出一种有效微调方法以强制满足计算律的组合性。大量实证研究表明，更好遵循计算律能持续提升多个基准上的推理性能，并揭示性质与规律间的协同效应。
原文摘要 · Abstract (English)
Despite the superior performance of Large Reasoning Models (LRMs), their reasoning behaviors are often counterintuitive, leading to suboptimal reasoning capabilities. To theoretically formalize the desired reasoning behaviors, this paper presents the Laws of Reasoning (LoRe), a unified framework that characterizes intrinsic reasoning patterns in LRMs. We first propose compute law with the hypothesis that the reasoning compute should scale linearly with question complexity. Beyond compute, we extend LoRe with a supplementary accuracy law. Since the question complexity is difficult to quantify in practice, we examine these hypotheses by two properties of the laws, monotonicity and compositionality. We therefore introduce LoRe-Bench, a benchmark that systematically measures these two tractable properties for large reasoning models. Evaluation shows that most reasoning models exhibit reasonable monotonicity but lack compositionality. In response, we develop an effective finetuning approach that enforces compute-law compositionality. Extensive empirical studies demonstrate that better compliance with compute laws yields consistently improved reasoning performance on multiple benchmarks, and uncovers synergistic effects across properties and laws. Project page: https://lore-project.github.io/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。