Back to News
Harvard University and Stanford University•February 4, 2026
OAT: Ordered Action Tokenization
Open Source
Robotics
RL
### TL;DR
OAT is a learned action tokenizer designed to enhance autoregressive policies in robot learning by discretizing continuous action sequences into ordered tokens. It addresses challenges in action tokenization by providing reasonable compression, universal decodability, and a left-to-right causally ordered token space.
Key Insights & Metrics
Pricing
Free
Cost structure
Version
1.0
Current release version
Hardware
CPU only
Compute requirements
Category
Open Source
Licensing model
Region
United States
Primary region
Key Features
- Transformer-based register tokens
- Finite scalar quantization
- Ordering-inducing training mechanisms
Discussion
0
Upvotes
0
Downvotes
0 reviews
Sign in to leave a review
Reviews
No reviews yet. Be the first to review!