Paper
Intermediate
The BERT Paper
Devlin 等
Overview
The classic paper that introduced bidirectional pretraining, establishing the pretrain-then-finetune paradigm in NLP.
https://arxiv.org/abs/1810.04805
Related resources
View all Large Language Models →
Official Course
Intermediate
Stanford CS224n Natural Language Processing with Deep Learning
斯坦福大学 · Large Language Models
A core NLP course running from word vectors and RNNs up to Transformers and large models, always tracking the frontier, with all materials public.
Official Course
Intermediate
★ Featured
Hugging Face NLP Course
Hugging Face · Large Language Models
Hugging Face's official course covering Transformer architectures, fine-tuning, and the Transformers ecosystem. Completely free and widely recommended.
Official Course
Intermediate
Hugging Face LLM Course
Hugging Face · Large Language Models
A systematic walkthrough of LLM architecture, tokenization, pretraining, alignment, and inference optimization, with runnable experiments. All free.
Official Docs
Intermediate
★ Featured
OpenAI API Official Docs
OpenAI · Large Language Models
OpenAI's official API documentation covering model capabilities, function calling, structured outputs, and best practices. The primary source for API-driven development.
Official Docs
Intermediate
OpenAI Cookbook
OpenAI · Large Language Models
OpenAI's official collection of runnable examples for common scenarios such as RAG, Agents, and function calling. Learn by copying.
Official Docs
Intermediate
Anthropic Claude Official Docs
Anthropic · Large Language Models
Official Claude API documentation covering the Messages API, tool use, MCP, and prompt engineering. Exceptionally well written.