“There are great tokenizers for text and images, but existing action tokenizers don’t…”
Physical Intelligence @physical_intX
Archived text
There are great tokenizers for text and images, but existing action tokenizers don’t work well for dexterous, high-frequency control. We’re excited to release (and open-source) FAST, an efficient tokenizer for robot actions.
With FAST, we can train dexterous generalist policies via simple next token prediction, and get a 5x training speed-up over prior state of the art!
Cited in
- Modelπ0-FAST