Deep Learning18 min read
Transformer 架构与训练详解
从整体架构、注意力机制到优化器与大规模预训练，系统梳理 Transformer 的组成和训练过程。
Writing
Blog
Formalized essays and longer reflections on the questions that shape my research practice.
Reinforcement Learning1 min read
Bellman Optimality as a Planning Lens
A compact note on why the Bellman equation is more than a recurrence: it is a way to organize decisions under uncertainty.
World Model1 min read
What I Mean by a World Model
A working definition for connecting perception, temporal prediction, and action in embodied learning systems.