课程资源
推荐阅读
按讲次组织的核心阅读清单与延伸材料
本列表在学期进行中持续更新。每讲的完整参考文献在该讲的 references 子页。
通用基础
- Jurafsky & Martin, Speech and Language Processing(第 3 版草稿免费在线)
- 李宏毅《生成式 AI 导论》公开课
- Andrej Karpathy, Neural Networks: Zero to Hero 系列
第 1-2 讲 · 经典 NLP 与 Transformer
- Attention Is All You Need(Vaswani et al., 2017)
- The Illustrated Transformer(Jay Alammar)
- The Annotated Transformer(Harvard NLP)
- Hugging Face NLP Course(chapters 1-6)
第 3 讲 · 提示词
- Chain-of-Thought Prompting(Wei et al., 2022)
- Tree of Thoughts(Yao et al., 2023)
- DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines(Khattab et al., 2023)
- TextGrad: Automatic "Differentiation" via Text(Yuksekgonul et al., 2024)
第 4 讲 · RAG
- Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks(Lewis et al., 2020)
- Retrieval-Augmented Generation for LLMs: A Survey(Gao et al., 2024)
- Self-RAG(Asai et al., 2023)
- LlamaIndex / LangChain 官方文档
第 5-6 讲 · 数据工程与微调
- Self-Instruct(Wang et al., 2022)
- MAGPIE(Xu et al., 2024)
- LoRA: Low-Rank Adaptation of LLMs(Hu et al., 2021)
- QLoRA(Dettmers et al., 2023)
- Tülu 3(AI2, 2024)
第 7 讲 · Agent
- ReAct(Yao et al., 2022)
- AutoGen(Wu et al., 2023)
- The Landscape of Emerging AI Agent Architectures
- Anthropic, Building Effective Agents(官方博客)
第 8 讲 · LLM as Judge
- Judging LLM-as-a-Judge with MT-Bench(Zheng et al., 2023)
- Prometheus 2(Kim et al., 2024)
- A Survey on LLM-as-a-Judge(Gu et al., 2024)
第 9 讲 · GEO
- GEO: Generative Engine Optimization(Aggarwal et al., 2023)
- Perplexity Labs 相关博客