FAST: Efficient Action Tokenization for Vision-Language-Action Models

This is quite important.

How do robot tokenizers work?

Citation

(Pertsch et al., 2025)

Pertsch, K., Stachowicz, K., Ichter, B., Driess, D., Nair, S., Vuong, Q., Mees, O., Finn, C., & Levine, S. (2025). FAST: Efficient Action Tokenization for Vision-Language-Action Models. https://arxiv.org/pdf/2501.09747