This repository provides a comprehensive textbook on the foundations of Large Language Models (LLMs), aimed at students, researchers, and practitioners interested in the field. It offers a systematic introduction to core concepts and cutting-edge techniques, with a focus on readability, rigor, and depth.

How It Works

The book systematically covers key LLM topics, including traditional language models, LLM architecture evolution, prompt engineering, parameter-efficient fine-tuning, model editing, and retrieval-augmented generation. Each chapter uses an animal analogy for illustrative purposes, enhancing accessibility. The content is derived from the authors' research and understanding, with a commitment to monthly updates and incorporating community feedback.

Quick Start & Requirements

The complete PDF version of the book is available as 大模型基础.pdf. Chapter-specific PDFs are located in the 大语言模型分章节内容 folder, and related papers are in the 大语言模型相关论文 folder.

Highlighted Details

• Covers six core chapters: Language Model Basics, Large Language Models, Prompt Engineering, Parameter-Efficient Fine-Tuning, Model Editing, and Retrieval-Augmented Generation.
• Includes paper lists for each chapter to track the latest advancements.
• Features animal-themed illustrations for each chapter to aid comprehension.
• Plans to expand coverage to include LLM inference acceleration and LLM agents in future versions.

Maintenance & Community

The project is actively maintained with a commitment to monthly updates. Feedback and suggestions are encouraged via GitHub issues. Contact is available via email at xuwenyi@zju.edu.cn.

Licensing & Compatibility

The repository content is not explicitly licensed. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

The book is presented as a first edition, with ongoing development and potential for revisions based on community input. Specific technical prerequisites or compatibility notes for accessing or utilizing the content are not detailed.