跳转至

知识图谱上的神经符号推理：基于查询视角的全景综述

文章背景与核心概要

知识图谱推理是人工智能、数据挖掘以及网络科学等领域的核心技术，它通过利用人类知识的结构化宝库来推断新信息。然而，传统的符号推理在面对图谱中不完整和含有噪声的数据时往往显得力不从心。为此，神经符号人工智能（Neural-Symbolic AI）应运而生，它有机结合了深度学习的鲁棒性与符号推理的精确性，致力于构建兼具高可解释性与通用性的AI系统。

本文献是一篇关于知识图谱推理的权威综述。文章创新性地从查询类型和神经符号方法分类的独特视角出发，深入探讨了如何通过融合深度学习与符号推理来克服传统局限。此外，该综述还重点关注了现代大语言模型（LLMs）在知识图谱推理工作流中的前沿应用，展示了其在知识提取与合成方面开辟的全新方向。

执行摘要 (Executive Summary)

本论文对知识图谱（KG）推理进行了全面综述，从查询类型和神经符号方法分类的视角对该领域进行了深入剖析。文章探讨了将深度学习与符号推理相结合如何克服传统方法在处理不完整和噪声数据方面的局限性，同时突出了现代大语言模型（LLMs）在KG推理工作流中的融合应用。

This paper provides a comprehensive survey of knowledge graph (KG) reasoning, examining the field through the lens of query types and neural-symbolic methodology classification. It explores how merging deep learning with symbolic reasoning overcomes traditional limitations regarding incomplete and noisy data, while also highlighting the integration of modern Large Language Models (LLMs) into KG reasoning workflows.

文档元数据 (Document Metadata)

字段 详情
标题 知识图谱上的神经符号推理：基于查询视角的全景综述 (Neural-Symbolic Reasoning over Knowledge Graphs: A Survey from a Query Perspective)
作者 Lihui Liu, Zihao Wang, Hanghang Tong
主要学科 人工智能 (cs.AI)
arXiv 标识符 arXiv:2412.10390
DOI 10.48550/arXiv.2412.10390
提交历史 • v1: 2024年11月30日
• v2: 2026年8月22日（当前版本）
Field Detail
Title Neural-Symbolic Reasoning over Knowledge Graphs: A Survey from a Query Perspective
Authors Lihui Liu, Zihao Wang, Hanghang Tong
Primary Subject Artificial Intelligence (cs.AI)
arXiv Identifier arXiv:2412.10390
DOI 10.48550/arXiv.2412.10390
Submission History • v1: 30 Nov 2024
• v2: 22 Aug 2026 (Current)

摘要 (Abstract)

知识图谱推理在数据挖掘、人工智能、万维网以及社会科学等多个领域中至关重要。这些知识图谱充当了人类知识的综合知识库，促进了新信息的推理。

尽管传统的符号推理具有其优势，但它在应对这些图表中不完整和含噪声数据所带来的挑战时显得力不从心。相比之下，神经符号人工智能（Neural Symbolic AI）的兴起标志着一项重大进展，它将深度学习的鲁棒性与符号推理的精确性结合起来。这种集成旨在开发出不仅高度可解释和可阐明，而且具备通用性的AI系统，从而有效地弥合符号方法与神经方法之间的鸿沟。

此外，大语言模型（LLMs）的出现为知识图谱推理开辟了新的前沿，实现了前所未有的知识提取与合成方式。本综述对知识图谱推理进行了深入回顾，重点关注各类查询类型以及神经符号推理的分类，并探讨了KG与LLMs的创新性融合。

Knowledge graph reasoning is pivotal in various domains such as data mining, artificial intelligence, the Web, and social sciences. These knowledge graphs function as comprehensive repositories of human knowledge, facilitating the inference of new information.

Traditional symbolic reasoning, despite its strengths, struggles with the challenges posed by incomplete and noisy data within these graphs. In contrast, the rise of Neural Symbolic AI marks a significant advancement, merging the robustness of deep learning with the precision of symbolic reasoning. This integration aims to develop AI systems that are not only highly interpretable and explainable but also versatile, effectively bridging the gap between symbolic and neural methodologies.

Additionally, the advent of large language models (LLMs) has opened new frontiers in knowledge graph reasoning, enabling the extraction and synthesis of knowledge in unprecedented ways. This survey offers a thorough review of knowledge graph reasoning, focusing on various query types and the classification of neural symbolic reasoning, alongside the innovative integration of KGs with LLMs.

访问与资源 (Access & Resources)