脑启发两阶段方法：通过模仿人类思维过程增强数学推理
Brain-Inspired Two-Stage Approach: Enhancing Mathematical Reasoning by Imitating Human Thought Processes
AI总结：
提出脑启发两阶段方法 Brain，由额叶模型制定计划、顶叶模型生成并执行代码，从而增强多步数学推理，并在 Code LLaMA 7B 基线上达到 SOTA。
摘要：
尽管大型语言模型在求解数学应用题方面展现出涌现能力，但在复杂的多步数学推理任务中仍存在一项具有挑战性的任务。为提升模型在数学推理任务上的性能，先前工作通过改进数据质量和数量，对开源模型进行了监督微调。本文提出一种名为 Brain 的新颖方法，通过模仿人类思维过程来增强数学推理能力：使用 Frontal Lobe Model（额叶模型）生成计划，随后采用 Parietal Lobe Model（顶叶模型）生成代码并执行以获得答案。首先，通过该方法，与基于 Code LLaMA 7B 的模型相比，我们取得了 SOTA 性能。其次，我们发现计划可以从自然语言、代码或形式语言中被显式提取。我们的代码和数据公开发布于 https://github.com/cyzhh/Brain。
英文摘要：
Although large language models demonstrate emergent abilities in solving math word problems, there is a challenging task in complex multi-step mathematical reasoning tasks. To improve model performance on mathematical reasoning tasks, previous work has conducted supervised fine-tuning on open-source models by improving the quality and quantity of data. In this paper, we propose a novel approach, named Brain, to imitate human thought processes to enhance mathematical reasoning abilities, using the Frontal Lobe Model to generate plans, and then employing the Parietal Lobe Model to generate code and execute to obtain answers. First, we achieve SOTA performance in comparison with Code LLaMA 7B based models through this method. Secondly, we find that plans can be explicitly extracted from natural language, code, or formal language. Our code and data are publicly available at https://github.com/cyzhh/Brain.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
