《大模型基础》：中英双语开放教材与论文路线图
ZJU-LLMs 的系统性入门教材从 n-gram、RNN、Transformer 与采样评测出发，覆盖大模型架构、Prompt 工程、参数高效微调、模型编辑和 RAG。仓库同时提供中文 290 页完整书、英文 368 页 draft、六个中文分章 PDF，以及按 31 个主题组织的 238 条论文与代码清单；适合用作 LLM 知识地图，再配合实现型课程练习。英文版由 GPT 直接翻译且仍待润色，核心教材文件也停留在 2024 年末，阅读近期进展时需另行核对。
Foundations of LLMs: A Bilingual Open Textbook and Paper Roadmap
ZJU-LLMs builds a structured introduction from n-grams, RNNs, Transformers, sampling, and evaluation through LLM architectures, prompt engineering, parameter-efficient finetuning, model editing, and RAG. The repository includes a 290-page Chinese book, a 368-page draft English translation, chapter PDFs, and a 238-entry paper/code reading list across 31 topics. It is best used as a foundational map alongside hands-on implementation resources and newer primary sources.
https://github.com/ZJU-LLMs/Foundations-of-LLMs
GitHub
GitHub - ZJU-LLMs/Foundations-of-LLMs: A book for Learning the Foundations of LLMs
A book for Learning the Foundations of LLMs. Contribute to ZJU-LLMs/Foundations-of-LLMs development by creating an account on GitHub.
August 25, 2026 10 1
