KG-CF: Knowledge Graph Completion with Context Filtering under the Guidance of Large Language Models

TL;DR

This paper introduces KG-CF, a framework that enhances knowledge graph completion by using large language models to filter relevant contexts, improving ranking accuracy in real-world datasets.

Contribution

KG-CF is a novel framework that applies LLMs for context filtering in ranking-based knowledge graph completion tasks, addressing redundancy and practical application challenges.

Findings

KG-CF outperforms existing methods on real-world datasets.

LLMs effectively filter irrelevant contexts, boosting completion accuracy.

Framework demonstrates practical utility in knowledge graph tasks.

Abstract

Large Language Models (LLMs) have shown impressive performance in various tasks, including knowledge graph completion (KGC). However, current studies mostly apply LLMs to classification tasks, like identifying missing triplets, rather than ranking-based tasks, where the model ranks candidate entities based on plausibility. This focus limits the practical use of LLMs in KGC, as real-world applications prioritize highly plausible triplets. Additionally, while graph paths can help infer the existence of missing triplets and improve completion accuracy, they often contain redundant information. To address these issues, we propose KG-CF, a framework tailored for ranking-based KGC tasks. KG-CF leverages LLMs' reasoning abilities to filter out irrelevant contexts, achieving superior results on real-world datasets. The code and datasets are available at \url{https://anonymous.4open.science/r/KG-CF}.