Milestone

vLLM PagedAttention

gpt-oss 源码/模型分析

AI Agents

《Agentic Design Pattern》学习

Code Agent

Claude Code

Code X

Deep Research

通义 Deep Research、

Tongyi DeepResearch

Search-r1

R1-Searcher

Agentic RL

Policy Optimization Algorithm

• 只从代码、论文、实验结果来学习

GRPO有什么问题？

GSPO

DAPO

Tool-integrated Reasoning

• multi turn tool use
• multi tools
• 同步调用tool还是异步？
• 怎么选择tool？

TORA

ToRL

Frameworks

Verl

VerlTool

LangChain LangGraph Agent总结

LangChain 03：Message解析

Multi-Agent System Plan & Execute