让大模型像按地图走一样推理，提升稳定性和准确性。
From Implicit Exploration to Structured Reasoning: Leveraging Guideline and Refinement for LLMs
- 用成功与失败案例提取结构化推理规则作为导航指南。
- 每步推理后自动纠错，使过程更稳定、结果更可靠。
- 适合需要高精度推理的场景，如数学题和编程任务。
大语言模型在通用推理任务中表现优异，但现有方法多依赖随机无序的隐式探索——如同无地图行走，导致推理路径不稳定、无法纠错且难以积累经验。为此，本文提出从隐式探索转向结构化推理的新框架，通过提取成功轨迹中的结构化模式与失败信号中的反思信息构建指导规则。推理时，模型分步遵循这些规则，并在每步后进行修正以纠正错误、稳定过程。在BBH及四个额外基准（GSM8K、MATH-500、MBPP、HumanEval）上的实验表明，该方法在多种推理任务中持续优于强基线。结构化推理结合逐步执行与修正机制，显著提升稳定性与泛化能力；同时，指导规则具备跨领域迁移性，支持跨模型协作，其效果可媲美甚至超越监督微调，在可扩展性上更具优势。
原文摘要 · Abstract (English)
Large language models (LLMs) have advanced general-purpose reasoning, showing strong performance across diverse tasks. However, existing methods often rely on implicit exploration, where the model follows stochastic and unguided reasoning paths-like walking without a map. This leads to unstable reasoning paths, lack of error correction, and limited learning from past experience. To address these issues, we propose a framework that shifts from implicit exploration to structured reasoning through guideline and refinement. First, we extract structured reasoning patterns from successful trajectories and reflective signals from failures. During inference, the model follows these guidelines step-by-step, with refinement applied after each step to correct errors and stabilize the reasoning process. Experiments on BBH and four additional benchmarks (GSM8K, MATH-500, MBPP, HumanEval) show that our method consistently outperforms strong baselines across diverse reasoning tasks. Structured reasoning with stepwise execution and refinement improves stability and generalization, while guidelines transfer well across domains and flexibly support cross-model collaboration, matching or surpassing supervised fine-tuning in effectiveness and scalability.
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。