GS-KGC: A Generative Subgraph-based Framework for Knowledge Graph Completion with Large Language Models

TL;DR

This paper introduces GS-KGC, a novel framework that leverages subgraph information and a QA approach to improve knowledge graph completion using large language models, achieving significant accuracy gains.

Contribution

The paper presents a new subgraph-based framework for KGC that enhances LLM reasoning with subgraph context and a QA approach, outperforming existing LLM-based methods.

Findings

GS-KGC achieves a 5.6% increase in Hits@3 on FB15k-237N.

GS-KGC outperforms previous models by 9.3% on ICEWS14.

The framework effectively discovers new triples and facts beyond existing KGs.

Abstract

Knowledge graph completion (KGC) focuses on identifying missing triples in a knowledge graph (KG) , which is crucial for many downstream applications. Given the rapid development of large language models (LLMs), some LLM-based methods are proposed for KGC task. However, most of them focus on prompt engineering while overlooking the fact that finer-grained subgraph information can aid LLMs in generating more accurate answers. In this paper, we propose a novel completion framework called \textbf{G}enerative \textbf{S}ubgraph-based KGC (GS-KGC), which utilizes subgraph information as contextual reasoning and employs a QA approach to achieve the KGC task. This framework primarily includes a subgraph partitioning algorithm designed to generate negatives and neighbors. Specifically, negatives can encourage LLMs to generate a broader range of answers, while neighbors provide additional contextual insights for LLM reasoning. Furthermore, we found that GS-KGC can discover potential triples within the KGs and new facts beyond the KGs. Experiments conducted on four common KGC datasets highlight the advantages of the proposed GS-KGC, e.g., it shows a 5.6\% increase in Hits@3 compared to the LLM-based model CP-KGC on the FB15k-237N, and a 9.3\% increase over the LLM-based model TECHS on the ICEWS14.