面向有强化学习基础的 Machine Learning PhD：从 world model、embodied imitation 到 VLA、action representation 与最新 VLA+world model 融合趋势。
This series is written for readers with reinforcement learning and machine learning background who want a structured path into embodied policies, VLA models, and the emerging VLA + world model frontier.
How to Build a Robot Foundation Model — Routes and Roadblocks
A first-principles chain: what a robot foundation model is, what it is made of, why the machine that produced large language models does not transfer, what that failure costs in each subsystem, and the order in which to build it anyway.
How to Build a Robot Foundation Model — Routes and Roadblocks (TLDR)
The deck edition: the same argument as one slide per screen, readable in fifteen minutes instead of fifty.
How to Build Generalist Physical AI: π0.7 from Beginner to Pro
A reproduction-grade walkthrough of π0.7 — every disclosed hyperparameter, every subsystem’s I/O contract, and an honest ledger of what the paper does not tell you.