From Symbolic to Neural and Back: Exploring Knowledge Graph-Large Language Model Synergies

TL;DR

This survey explores the integration of Knowledge Graphs and Large Language Models, highlighting approaches, challenges, and future directions to enhance reasoning, reduce hallucinations, and improve knowledge management in AI systems.

Contribution

It systematically categorizes existing methods of KG-LLM integration, emphasizing scalability, efficiency, and data quality, and proposes future research directions.

Findings

KG-enhanced LLMs improve reasoning and reduce hallucinations

LLM-augmented KGs facilitate construction and querying

Identifies gaps and future directions in knowledge integration

Abstract

Integrating structured knowledge from Knowledge Graphs (KGs) into Large Language Models (LLMs) enhances factual grounding and reasoning capabilities. This survey paper systematically examines the synergy between KGs and LLMs, categorizing existing approaches into two main groups: KG-enhanced LLMs, which improve reasoning, reduce hallucinations, and enable complex question answering; and LLM-augmented KGs, which facilitate KG construction, completion, and querying. Through comprehensive analysis, we identify critical gaps and highlight the mutual benefits of structured knowledge integration. Compared to existing surveys, our study uniquely emphasizes scalability, computational efficiency, and data quality. Finally, we propose future research directions, including neuro-symbolic integration, dynamic KG updating, data reliability, and ethical considerations, paving the way for intelligent systems capable of managing more complex real-world knowledge tasks.