Natural Language Processing

An Approach Based on Large Language Models
Erscheinungsjahr 2027
ISBN: 978-981-9288-26-7
Verlag: Springer

Buch, Englisch, Format (B × H): 155 mm x 235 mm

Natural Language Processing (NLP) is often regarded as a cornerstone of artificial intelligence. The advent of deep learning has revolutionized NLP techniques, particularly with the rise of large language models in recent years, which have established a new paradigm for NLP research. This book introduces fundamental concepts in natural language processing and deep learning while emphasizing modern NLP techniques driven by pre-trained and large language models.

The book is structured into three main sections:

• Foundational Knowledge: This section covers the essential principles of natural language processing and deep learning, including key toolsets and commonly used datasets.
• Pre-trained Language Models: This section explores language models, pre-trained word embeddings, methodologies for implementing pre-trained models, and their applications.
• Large Language Models: This section begins with an overview of pre-training methods for large language models, followed by discussions on adaptation, application, and evaluation strategies. It also examines extension techniques inspired by pre-trained models and concludes with an illustrative example using the DeepSeek series of models, showcasing the latest advancements in large language models.

Beyond theoretical foundations, this book incorporates practical case studies with accompanying PyTorch code implementations, enabling readers to deepen their understanding while seamlessly applying NLP models in practice—bridging theory with hands-on experience.

Designed for college students, academic researchers, and computer engineers with a basic knowledge of machine learning, this book serves as a valuable resource for those seeking to advance their expertise in NLP algorithms. It is also an essential reference for students and professionals interested in artificial intelligence, deep learning, large language models, and natural language processing, as well as individuals aspiring to enter the field of AI applications.

Zielgruppe

Research

Autoren/Hrsg.

Che, Wanxiang

Guo, Jiang

Cui, Yiming

Fachgebiete

• Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz Maschinelles Lernen
• Mathematik | Informatik EDV | Informatik Programmierung | Softwareentwicklung Programmier- und Skriptsprachen
• Geisteswissenschaften Sprachwissenschaft Computerlinguistik, Korpuslinguistik
• Mathematik | Informatik EDV | Informatik Informatik Natürliche Sprachen & Maschinelle Übersetzung

Inhaltsverzeichnis

Part I. Foundations.- Chapter 1. Introduction.- Chapter 2. Foundations of Natural Language Processing.- Chapter 3. Basic Toolkits and Common Datasets.- Chapter 4. Neural Network Foundations for Natural Language Processing.- Part II. Pretrained Language Models.- Chapter 5. Language Models.- Chapter 6. Pretrained Word Embeddings.- Chapter 7. Pretrained Language Models.- Part III. Large Language Models.- Chapter 8. Pretraining of Large Language Models.- Chapter 9. Adaptation of Large Language Models.- Chapter 10. Applications of Large Language Models.- Chapter 11. Evaluation of Large Language Model Capabilities.- Chapter 12. Extensions of Pretrained Language Models.- Chapter 13. An Introduction to the Principles of the DeepSeek Family.- References.- Terminology.

Über Autor(innen)

Wanxiang Che, full Professor and PhD Supervisor at the School of Computing, Harbin Institute of Technology (HIT), Vice Dean of the Artificial Intelligence Research Institute, National Youth Talent, Young Scholar of Longjiang Scholars Program, and Visiting Scholar at Stanford University. He currently serves as a Council Member of the Chinese Society for Computational Linguistics (CSCWL), Deputy Director and Secretary-General of the Professional Committee on Computational Linguistics; Executive Board Member and Secretary-General of the Asia-Pacific Chapter of the Association for Computational Linguistics (AACL); and Co-Chair of the Program Committee for ACL 2025. He has undertaken multiple research projects, including key projects and special projects from the National Natural Science Foundation of China, as well as sub-projects under the "Next Generation Artificial Intelligence" initiative in the 2030 Major Project. He has received numerous awards, including a nomination for the Best Paper Award at AAAI 2013, the First Prize for Scientific and Technological Progress in Heilongjiang Province, and the Heilongjiang Provincial Youth Science and Technology Award.

Jiang Guo, currently engaged in research on large models, retrieval-enhanced generation technology, and enterprise data intelligence, with a focus on the application and development of these technologies in cloud services. Previously served as a Postdoctoral Researcher at the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT, and holds a PhD from Harbin Institute of Technology, having also been jointly trained at Johns Hopkins University. In the field of artificial intelligence and natural language processing, he has published over 30 academic papers, which have been cited more than 2,400 times (according to Google Scholar statistics), and has authored one academic monograph and one translation. He has received the Baidu Scholarship and was nominated for the CSCWL's "Outstanding Doctoral Dissertation" award in 2018. He has been invited multiple times to serve as a Senior Program Committee Member and Area Chair for top international academic conferences.

Yiming Cui, Senior Engineer, currently serving as a Senior Scientist at iFlytek Research Institute and Vice Dean of the Beijing Research Institute of iFlytek. He is an IEEE Senior Member and a CCF Senior Member. He holds a PhD from Harbin Institute of Technology. His main research areas include large models, pre-trained models, and machine reading comprehension, where he has won over 20 championships in international evaluations in related fields. The Chinese pre-trained models and large model open-source projects he led, such as Chinese-BERT-wwm and the Chinese-LLaMA-Alpaca series, have accumulated over 40,000 stars on open-source platforms, ranking among the top GitHub Trending global projects and developer rankings multiple times. He has published over 40 academic papers in conferences or journals such as ACL, EMNLP, AAAI, and IEEE/ACM TASLP, including one ESI highly cited paper. He has received a nomination for the Best Paper Award at SemEval-2022, and two of his papers were selected by Paper Digest as the most influential papers. He has been listed among the "Top 2% Global Scientists" multiple times. He serves as the Senior Area Chair for ACL 2025, Area Chair for EMNLP 2021 and AACL 2022, Action Editor for ARR, and Regular Reviewer for TACL, among other academic roles.