通过类比教学：教育类比生成的模块化流水线
Teaching Through Analogies: A Modular Pipeline for Educational Analogy Generation
AI总结：
提出一个模块化流水线，将教育类比生成分解为四个阶段，基于结构映射理论，评估12个LLM在两个数据集上的表现，发现子概念显著提升解释质量和封闭检索精度，并引入LLM作为评判的评估方法。
摘要：
类比通过将不熟悉的概念与已知概念联系起来，帮助学习者理解。尽管最近取得了进展，大型语言模型（LLM）在生成与人类质量相当的类比方面仍然困难。我们提出了一个用于教育类比生成的模块化流水线，将任务分解为四个阶段：源发现、子概念生成、解释生成和评估。基于结构映射理论，该流水线能够系统地、逐阶段地分析模型选择和输入配置如何影响类比质量。我们在两个具有结构化子概念注释的数据集（SCAR和ParallelPARC）上评估了来自六个模型家族的12个最先进的LLM，以及用于封闭设置检索的七个嵌入模型。我们的结果表明，子概念显著提高了解释质量和封闭设置检索精度，但在开放式源生成中提供的益处有限。我们进一步引入了一种LLM作为评判的评估方法，并针对七名注释者的人类注释验证了其评分，发现Claude Sonnet 4.6在人类排名上的对齐比细粒度绝对分数更可靠。综合来看，我们的发现揭示了孤立研究无法捕捉的跨阶段交互，并强调了子概念基础作为类比质量生成的关键驱动因素。
英文摘要：
Analogies help learners understand unfamiliar concepts by relating them to known concepts. Despite recent advances, large language models (LLMs) continue to struggle to generate analogies of comparable quality to those produced by humans. We present a modular pipeline for educational analogy generation, decomposing the task into four stages: source finding, sub-concept generation, explanation generation, and evaluation. Grounded in Structure Mapping Theory, the pipeline enables systematic, stage-by-stage analysis of how model choice and input configuration affect analogy quality. We evaluate 12 state-of-the-art LLMs across six model families on two datasets with structured sub-concept annotations (SCAR and ParallelPARC), alongside seven embedding models for closed-setting retrieval. Our results show that sub-concepts substantially improve explanation quality and closed setting retrieval precision but provide limited benefit in open-ended source generation. We further introduce an LLM-as-a-judge evaluation methodology and validate its scoring against human annotations from seven annotators, finding that Claude Sonnet 4.6 aligns more reliably with human rankings than with fine-grained absolute scores. Taken together, our findings reveal cross-stage interactions that isolated studies cannot capture, and highlight sub-concept grounding as a key driver of analogy quality generation.
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。
