让大模型通过多层自我反思自动优化数学推理过程
Advancing Multi-Step Mathematical Reasoning in Large Language Models through Multi-Layered Self-Reflection with Auto-Prompting
- 采用多层自省与自动生成提示，迭代修正推理错误
- 在多个基准上超越标准思维链，接近专用推理模型性能
- 适合希望提升通用模型推理能力的研究者和开发者
大型语言模型（LLMs）虽在解题能力上取得进展，但在复杂多步推理任务中仍表现不足。本文提出多层自省与自动生成提示（MAPS）框架，融合思维链（CoT）、自省机制与自动提示技术。该框架通过迭代优化：先以CoT生成解题路径，检测错误后，自省模块分析问题并生成针对性修正提示，引导模型逐步完善推理。在四个主流基准测试中，MAPS显著优于标准CoT，在多个LLM上表现优异，使通用模型达到与专用推理模型相当的水平。增加反思层数可提升准确率，但会增加计算开销；因此MAPS智能控制反思深度，实现成本与性能的最佳平衡。
原文摘要 · Abstract (English)
Recent advancements in Large Language Models (LLMs) have significantly improved their problem-solving capabilities. However, these models still struggle when faced with complex multi-step reasoning tasks. In this paper, we propose the Multi-Layered Self-Reflection with Auto-Prompting (MAPS) framework, a novel approach designed to enhance multi-step mathematical reasoning in LLMs by integrating techniques such as Chain of Thought (CoT), Self-Reflection, and Auto-Prompting. Unlike traditional static prompting methods, MAPS employs an iterative refinement process. Initially, the model generates a solution using CoT prompting. When errors are detected, an adaptive self-reflection mechanism identifies and analyzes them, generating tailored prompts to guide corrections. These dynamically adjusted prompts enable the model to iteratively refine its reasoning. Experiments on four well-established benchmarks across multiple LLMs show that MAPS significantly outperforms standard CoT and achieves competitive results with reasoning-optimized models. In addition, MAPS enables general-purpose LLMs to reach performance levels comparable to specialized reasoning models. While deeper reflection layers improve accuracy, they also increase token usage and costs. To balance this trade-off, MAPS strategically limits reflection depth, ensuring an optimal balance between cost and reasoning performance.
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。