Misha Lvovsky
commited on
Commit
·
99fe89b
1
Parent(s):
1fefe90
Update README.md to clarify the purpose of the fork and its modifications
Browse files
README.md
CHANGED
|
@@ -6,9 +6,12 @@ tags:
|
|
| 6 |
- tokenizer
|
| 7 |
---
|
| 8 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
# FAST: Efficient Action Tokenization for Vision-Language-Action Models
|
| 10 |
|
| 11 |
-
This is the official repo for the [FAST action tokenizer](https://www.pi.website/research/fast).
|
| 12 |
|
| 13 |
The action tokenizer maps any sequence of robot actions into a sequence of dense, discrete **action tokens** for training autoregressive VLA models.
|
| 14 |
|
|
|
|
| 6 |
- tokenizer
|
| 7 |
---
|
| 8 |
|
| 9 |
+
This is forked from [FAST action tokenizer](https://www.pi.website/research/fast).
|
| 10 |
+
|
| 11 |
+
This fork just adds a slight modification which pads the tokens to make it always output something rather than zeros as it does on errors in the original version.
|
| 12 |
+
|
| 13 |
# FAST: Efficient Action Tokenization for Vision-Language-Action Models
|
| 14 |
|
|
|
|
| 15 |
|
| 16 |
The action tokenizer maps any sequence of robot actions into a sequence of dense, discrete **action tokens** for training autoregressive VLA models.
|
| 17 |
|