Augmenting Large Language Models with Knowledge Graphs for Domain-Specific Applications

Sushanta Mohapatra, Tosin Adewumi

Keywords: Large language models, Knowledge graph, Retrieval augmented generation, Supply chain management

Abstract:

Large language models (LLMs) models demonstrate impressive capabilities in generating human-like text and handling general-purpose queries. However, their application in specialized domains, such as supply chain management (SCM), remains challenging due to limitations in understanding domain-specific terminology and concepts.
This research explores the integration of Knowledge Graphs (KGs) into Retrieval Augmented Generation (RAG) pipelines to enhance the performance of LLMs in domain-specific tasks. We introduce a novel benchmark dataset for SCM, covering eight supply chain functions and thirteen distinct categories of questions.
The results of this study demonstrated that the KG integration improved performance compared to traditional RAG approaches, with smaller models achieving notable gains that reduced the performance gap with larger models.

Serve As Reviewer: ~Tosin_Adewumi1

Submission Number: 21