Natural Language Processing: Neural Networks and Large Language Models by Tong Xiao, Jingbo Zhu instant download

Natural Language Processing: Neural Networks and Large Language Models by Tong Xiao, Jingbo Zhu instant download

This book focuses on modern NLP methods centered around neural networks and foundation models. It aims to provide a practical guide to understanding, building, and applying powerful models. The book is structured from the perspective of constructing neural NLP models and is divided into three main parts:

Part 1 (Chap. 1~2) introduces the core concepts and methods of machine learning and neural networks, laying a foundation for the subsequent chapters. Part 2 (Chap. 3~6) explains the neural networks used in NLP tasks, including word representation models, sequence models, and sequence-to-sequence models.  Transformers are introduced in this part with a dedicated chapter. Part 3 (Chap. 7~11), accounting for roughly half the entire book, focuses on large language models (LLMs), covering topics such as pretraining, generative models, prompt engineering, alignment, and inference. Advanced topics like very long sequence modeling, position extrapolation and interpolation, k-v caching, batching, state space models and high performance computing in training and inference are also discussed in details.

This treasure book, best of its kind perhaps, serves both as an introductory text for newcomers and as a reference manual for key concepts and methods in this field from which even more experienced researchers and practitioners can benefit.