The repository provides a large collection of hand-drawn SVG diagrams and PDF maps covering the architecture of LLMs and VLMs, training algorithms including RLHF, GRPO, DPO, SFT, and chain-of-thought distillation, and reinforcement learning fundamentals like policy gradient and PPO. It serves as a visual study guide complementing the published book 《大模型算法：强化学习、微调与对齐》. Topics span transformer structures, attention mechanisms, reward modeling, and RAG optimization.