用大模型把难懂的知识图谱规则变成人话，提升可读性。
Rule2Text: A Framework for Generating and Evaluating Natural Language Explanations of Knowledge Graph Rules
- 用大模型生成规则的自然语言解释，支持零样本、思维链等多种提示策略。
- 在多个数据集上验证效果，微调后解释质量显著提升，尤其在专业领域表现更好。
- 提供自动评估框架和高质量数据集，适合知识图谱可解释性研究者使用。
知识图谱可通过规则挖掘增强，但挖掘出的逻辑规则因复杂性和标注习惯差异，常难以被人理解。本文提出 Rule2Text 框架，利用大语言模型（LLMs）生成自然语言解释，提升知识图谱的可访问性与可用性。在 Freebase 变体（FB-CVT-REV、FB+CVT-REV、FB15k-237）及 ogbl-biokg 等数据集上，使用 AMIE 3.5.1 挖掘规则，系统评估多种 LLM 在零样本、少样本、类型信息融合及思维链推理等提示策略下的表现。通过人工评估验证生成解释的准确性和清晰度，并构建基于 LLM 作为裁判的自动化评估框架，其结果与人类评价高度一致。结合最佳模型（Gemini 2.0 Flash）、LLM 裁判与人机反馈，构建高质量真值数据集，用于微调开源模型 Zephyr。微调后解释质量显著提升，尤其在领域特定数据集上效果突出。此外，集成类型推断模块，支持缺乏显式类型信息的知识图谱。所有代码与数据已公开于 https://github.com/idirlab/KGRule2NL。
原文摘要 · Abstract (English)
Knowledge graphs (KGs) can be enhanced through rule mining; however, the resulting logical rules are often difficult for humans to interpret due to their inherent complexity and the idiosyncratic labeling conventions of individual KGs. This work presents Rule2Text, a comprehensive framework that leverages large language models (LLMs) to generate natural language explanations for mined logical rules, thereby improving KG accessibility and usability. We conduct extensive experiments using multiple datasets, including Freebase variants (FB-CVT-REV, FB+CVT-REV, and FB15k-237) as well as the ogbl-biokg dataset, with rules mined using AMIE 3.5.1. We systematically evaluate several LLMs across a comprehensive range of prompting strategies, including zero-shot, few-shot, variable type incorporation, and Chain-of-Thought reasoning. To systematically assess models' performance, we conduct a human evaluation of generated explanations on correctness and clarity. To address evaluation scalability, we develop and validate an LLM-as-a-judge framework that demonstrates strong agreement with human evaluators. Leveraging the best-performing model (Gemini 2.0 Flash), LLM judge, and human-in-the-loop feedback, we construct high-quality ground truth datasets, which we use to fine-tune the open-source Zephyr model. Our results demonstrate significant improvements in explanation quality after fine-tuning, with particularly strong gains in the domain-specific dataset. Additionally, we integrate a type inference module to support KGs lacking explicit type information. All code and data are publicly available at https://github.com/idirlab/KGRule2NL.
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。