MKGL: Mastery of a Three-Word Language

TL;DR

This paper introduces a specialized three-word language for knowledge graphs, enabling large language models to better understand and generate factual triplets with reduced errors and improved comprehension of unseen terms.

Contribution

The paper presents a novel three-word language framework for KGs, along with tailored learning techniques and context retrieval to improve LLM performance on KG tasks.

Findings

LLMs can fluently learn the three-word language with tailored methods.

Significant error reduction compared to traditional KG embedding methods.

Enhanced ability to generate and interpret three-word KG sentences.

Abstract

Large language models (LLMs) have significantly advanced performance across a spectrum of natural language processing (NLP) tasks. Yet, their application to knowledge graphs (KGs), which describe facts in the form of triplets and allow minimal hallucinations, remains an underexplored frontier. In this paper, we investigate the integration of LLMs with KGs by introducing a specialized KG Language (KGL), where a sentence precisely consists of an entity noun, a relation verb, and ends with another entity noun. Despite KGL's unfamiliar vocabulary to the LLM, we facilitate its learning through a tailored dictionary and illustrative sentences, and enhance context understanding via real-time KG context retrieval and KGL token embedding augmentation. Our results reveal that LLMs can achieve fluency in KGL, drastically reducing errors compared to conventional KG embedding methods on KG completion. Furthermore, our enhanced LLM shows exceptional competence in generating accurate three-word sentences from an initial entity and interpreting new unseen terms out of KGs.