Benchmark for Knowledge Graphs and LLM Accuracy

This technical report investigates the role of Knowledge Graphs in enhancing the accuracy of Large Language Models (LLMs) for question answering on enterprise SQL databases. It introduces a benchmark tailored to enterprise settings, revealing significant accuracy improvements when LLMs are used in conjunction with Knowledge Graphs. Key findings indicate that LLMs achieve a 16% accuracy rate on SQL databases, which rises to 54% with Knowledge Graph integration, highlighting the importance of busi