The Role Evolution of KGs in Synthesizing with LLMs: From Background Knowledge to Joint Reasoning

Abstract

This talk gives an overview of the recent works of synthesizing LLMs with KGs and highlights the evolving role of KGs, i.e., from KGs serving as passive background knowledge to the active involvement in joint reasoning processes with LLMs. It summarizes the strengths, limitations, and KG requirements of the approaches with different KG roles in synthesizing LLMs with KGs, and the applications of LLMs+KGs in question answering, recommendation, schema matching and data integration, system diagnostics, etc. It also discusses the open challenges and highlights the opportunities for LLMs+KGs.

Chuangtao Ma

Postdoctoral Researcher