|
Prof. Xin Wang
Tianjin University
Title: Knowledge Graph-Based Retrieval-Augmented Generation with Large Language Models
|
Abstract: Retrieval-Augmented Generation (RAG) has become a key technology for improving the accuracy of question answering with large language models (LLMs). This talk will first provide an overview of RAG methods for LLM-based question answering, and then introduce the research direction of knowledge graph-based retrieval augmentation (KG RAG), which targets the retrieval of interrelated factual information. Representative state-of-the-art works will be reviewed. Furthermore, the talk will explore representation learning techniques for knowledge graphs to enable semantic relevance-based retrieval in KG RAG. Finally, application case studies will be presented to demonstrate the practical effectiveness of KG RAG in vertical domains.
Short Biography: Xin Wang is a Distinguished Professor at the School of Artificial Intelligence at Tianjin University. Prof. Xin Wang's research interests include knowledge graphs, large language models, and knowledge data processing. He has been the Principal Investigator of the National Key Research and Development Project of China, and four research projects funded by the National Natural Science Foundation of China (NSFC). He has authored four books, published more than 150 research papers in various international conferences and journals, including IEEE TKDE, WWW, SIGMOD, VLDB, ICDE, IJCAI, AAAI, CIKM, and ISWC, and 20 engineering patents. He served as a Program Committee Chair of international conferences including WISE2025, DASFAA2023, and APWeb-WAIM2020. He is an Associate Editor of the international journals of Knowledge-Based Systems, World Wide Web, Data Science and Engineering, and Health Information Science and Systems.