用知识增强提示提升大模型解决比例类比题能力
KnowledgePrompts: Exploring the Abilities of Large Language Models to Solve Proportional Analogies via Knowledge-Enhanced Prompting
- 通过三类知识增强提示：范例、结构化知识和目标知识
- 最佳模型准确率仅55%，显示当前大模型仍难解类比问题
- 目标知识比其他知识形式更有效，适合认知推理研究者
类比是认知的基础。比例类比由四个词项组成，常用于评估语言与认知能力。例如，“Oxygen is to Gas as <blank> is to <blank>”需识别前两项的语义关系（如“种类”），并找到具有相同关系的第二对词项（如“Aluminum”和“Metal”）。本文构建了一个包含1.5万道多选题的类比完形数据集，并在多种知识增强提示设置下评估当代大语言模型的表现。具体地，我们通过范例、结构化知识和目标知识三种方式增强提示。结果显示，尽管经过大量训练，当前大模型在完成比例类比任务上仍面临挑战，最佳模型准确率为55%。值得注意的是，提供目标知识相比范例或结构化知识能更有效地帮助模型完成类比任务。代码与数据已开源：https://github.com/Thiliniiw/KnowledgePrompts/
原文摘要 · Abstract (English)
Making analogies is fundamental to cognition. Proportional analogies, which consist of four terms, are often used to assess linguistic and cognitive abilities. For instance, completing analogies like "Oxygen is to Gas as <blank> is to <blank>" requires identifying the semantic relationship (e.g., "type of") between the first pair of terms ("Oxygen" and "Gas") and finding a second pair that shares the same relationship (e.g., "Aluminum" and "Metal"). In this work, we introduce a 15K Multiple-Choice Question Answering (MCQA) dataset for proportional analogy completion and evaluate the performance of contemporary Large Language Models (LLMs) in various knowledge-enhanced prompt settings. Specifically, we augment prompts with three types of knowledge: exemplar, structured, and targeted. Our results show that despite extensive training data, solving proportional analogies remains challenging for current LLMs, with the best model achieving an accuracy of 55%. Notably, we find that providing targeted knowledge can better assist models in completing proportional analogies compared to providing exemplars or collections of structured knowledge. Our code and data are available at: https://github.com/Thiliniiw/KnowledgePrompts/
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。