Reliable Large Models with Knowledge Augmentation

Overview

Authors:

• Wenguang Wang ⁰

1. Wenguang Wang

   1. Shanghai, China
      

   View author publications

   Search author on: PubMed Google Scholar

In the rapidly evolving landscape of artificial intelligence, large language models (LLMs) have emerged as transformative tools, yet they face persistent challenges such as hallucination, knowledge obsolescence, and limited reasoning depth. The book "Knowledge-Enhanced Large Models" addresses these gaps head-on, offering a comprehensive roadmap for integrating structured knowledge systems—particularly knowledge graphs—with cutting-edge AI models to build more reliable, accurate, and context-aware intelligent systems. This book is tailored for AI researchers, data scientists, engineers, students, and practitioners seeking to harness the synergy between large models and knowledge representation technologies. It balances theoretical rigor with practical implementation, making it equally valuable for academic exploration and industrial application.

The book is organized into 10 chapters, systematically guiding readers from foundational concepts to advanced techniques and real-world applications. The first two chapters explore the rise of LLMs, their inherent limitations, and the paradigm shift toward knowledge-enhanced models. Chapters 3 to 5 delve into the infrastructure required to augment LLMs with structured knowledge. Chapters 6 to 9 explore cutting-edge methodologies for bridging symbolic knowledge systems with neural networks. The final chapter translates theory into practice, offering actionable guidelines for deploying knowledge-enhanced models across industries.

• Shanghai, China

  Wenguang Wang

Wang Wenguang, a distinguished expert in artificial intelligence, holds a Master's degree from Zhejiang University. Recognized as an Elite Talent of the Pudong New Area "Pearl Plan," he has received provincial and ministerial-level Science and Technology Progress Awards for his contributions to AI standard formulation. His accolades include serving as Chief Technician of Pudong New Area, being named Tencent Cloud Most Valuable Expert (TVP) and receiving the Outstanding Contributor award from the China Artificial Intelligence Industry Development Alliance.

As the author of the bestselling AI book Knowledge Graph: Cognitive Intelligence Theory and Practice, Mr. Wang is dedicated to advancing research and the practical application of general artificial intelligence technologies. His extensive professional engagements include membership in the Shanghai Artificial Intelligence Technology Standardization Committee and service as a review expert for the Shanghai Municipal Science and Technology Commission. He is also a senior member of the China Computer Federation (CCF) and plays key roles in several academic committees, including the Language and Knowledge Computing division of the Chinese Information Processing Society (CIPS) and the Deep Learning Specialized Committee of the Chinese Association for Artificial Intelligence (CAAI). Additionally, he is an expert committee member of the Shanghai Artificial Intelligence Technology Association.

He has made substantial contributions to the field, having played a key role in formulating over ten AI-related standards. His research portfolio includes numerous national invention patents and academic papers, and he has co-authored multiple authoritative books on artificial intelligence.

• Large Language Models
• Knowledge-Enhanced Models
• Knowledge Graphs
• Vector Databases
• Retrieval-Augmented Generation (RAG)
• Graph Databases
• GraphRAG
• Graph-Model Synergy Paradigm

Policies and ethics