š Welcome to Baltic Vectors
Hi there, my name is Shawn. I use this space to write down my learnings on anything and everything to do with AI and robotics.
What puts the A in VLA?
The field of robotics is experiencing an algorithmic shift towards VLAs (Vision Language Action) models - teaching transformers to āspeakā robot actions in the physical world. In this post, Iāll: Demystify robot actions - what do VLAs output to make robots move. Review how action representations have evolved over the past couple of years. I donāt have a robotics background so all of this is quite new to me, and Iāve always been perplexed by the bridging of what must happen between a model outputting tokens to an actual arm flipping a pancake. What do robot actions actually look like? ...