This weekend, I’m just going to read “RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control” and thought it might be more fun if a few people joined in.

I don’t have any robotics background myself; I’m just curious and trying to understand the paper as I go. If anyone is free and wants to drop by, you’re welcome to join. It would actually be better if people who are into robotics come, because I’ll probably have a lot of questions.

We’ll just go through the paper slowly, try to understand what it’s doing, and talk through anything that feels interesting or confusing along the way.

No need to read it beforehand; even just showing up and reading together is fine.