类比学习：通过基于计算图的检索增强数学应用题求解的少样本提示
Learning by Analogy: Enhancing Few-Shot Prompting for Math Word Problem Solving with Computational Graph-Based Retrieval
AI总结：
针对大语言模型（LLMs）解决数学应用题（MWPs）的推理难题，本文提出基于计算图检索相似问题作为示例的类比学习方法，在六个数据集上实现平均6.7%的绝对性能提升，有效增强LLMs的MWP求解能力。
摘要：
大语言模型（LLMs）在数学应用题（MWPs）等复杂推理任务中表现不佳。本文提出，通过类似结构问题的类比可提升LLMs的MWP求解能力。具体而言，我们检索与给定问题具有相似计算图的问题作为提示中的示例，为生成模型提供可参考的正确推理路径。在六个数学应用题数据集上的实证结果表明，所提方法有效，与基线方法相比，平均绝对提升高达6.7%。这些结果凸显了该方法在解决当前LLMs推理挑战方面的潜力。
英文摘要：
Large language models (LLMs) are known to struggle with complicated reasoning tasks such as math word problems (MWPs). In this paper, we present how analogy from similarly structured questions can improve LLMs' problem-solving capabilities for MWPs. Specifically, we rely on the retrieval of problems with similar computational graphs to the given question to serve as exemplars in the prompt, providing the correct reasoning path for the generation model to refer to. Empirical results across six math word problem datasets demonstrate the effectiveness of our proposed method, which achieves a significant improvement of up to 6.7 percent on average in absolute value, compared to baseline methods. These results highlight our method's potential in addressing the reasoning challenges in current LLMs.
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。
