FAST: Frequency-space Action Sequence Tokenization

8 min read

Turn a smooth, high-frequency robot trajectory into a compact motion sentence that an autoregressive VLA can predict.

InputAction chunk A ∈ ℝᴴˣᴰTransformDCT over timeQuantization scaleγ = 10Learned partBPE vocabulary onlyFAST vocabulary1,024 tokensFAST+ corpus~1M action chunksVLA lossCross-entropy

why FAST

Stop making the model predict the same motion 700 times

Per-timestep binning preserves temporal redundancy. FAST compresses the whole trajectory before the Transformer sees it.

Scalar binning

0.31, 0.32, 0.32, 0.33 → 124, 124, 124, 125

One token per dimension per timestep. A 50 Hz, 14-dimensional, one-second chunk becomes about 700 tokens, and copying the previous token can be an easy shortcut.

FAST

smooth trajectory → sparse frequency coefficients

Describe the whole chunk in frequency space, then compress recurring patterns. On the paper's 50 Hz shirt-folding data, roughly 700 scalars became 53 tokens, a 13.2× reduction.

encode

Action chunk → compact token sequence

Only rounding loses information; BPE can exactly recover the quantized coefficient stream.

01

Normalize

Map each dataset/action dimension's 1st and 99th percentiles to approximately [−1, 1].

A → Ã
02

DCT over time

Rewrite each dimension as average motion, slow shape, then progressively faster corrections.

C = DCT(Ã)
03

Scale + round

Quantize continuous coefficients. Tiny high-frequency terms become zero, creating a sparse matrix.

C̄ = round(γC)
04

Frequency first

Place frequency 0 for every action dimension first, then frequency 1, and so on.

[C̄₁,₀ … C̄ᴰ,₀, C̄₁,₁ …]
05

BPE compression

Merge frequent neighboring integer patterns into tokens from a compact fitted vocabulary.

[12, 0, 0, 0] → T₄₂

A FAST token has no fixed meaning like “one joint” or “one timestep.” It may encode several neighboring coefficients, dimensions, or frequencies, depending on which patterns BPE found in the training corpus.

ordering

Broad motion first, fine corrections later

Frequency-major flattening makes the autoregressive sequence progressively refine the complete multi-dimensional trajectory.

Prediction order
dim 1
dim 2
dim D
frequency 0
overall
overall
overall
frequency 1
slow
slow
slow
frequency 2
medium
medium
medium
later frequencies
fine
fine
fine

tokenizer fitting

FAST is mostly fixed math, not a neural encoder

There is no learned encoder, decoder, reconstruction loss, or codebook loss inside the tokenizer.

Fixed operations

Percentile normalization
DCT and inverse DCT
Scaling and rounding
Frequency-major flattening

The learned part

One BPE merge dictionary

Count adjacent pairs in quantized action sequences, merge the most frequent pair, add a vocabulary entry, and repeat. FAST+ fits one shared dictionary on about one million one-second chunks from diverse robots and action spaces, padding action vectors to 32 dimensions so different embodiments can share the corpus.

This fitting is frequency counting and pair merging, not backpropagation. Images, instructions, and cross-entropy only enter later, when a VLA learns what the fitted tokens mean.

decode

Reverse the pipeline to recover continuous actions

An autoregressive VLA such as π₀-FAST generates tokens one at a time, then decodes the entire predicted chunk.

FAST tokens

T₁ … Tₙ

BPE decode

integer stream

Unflatten

C̄ ∈ ℤᴰˣᴴ

Divide by γ

Ĉ = C̄ / γ

Inverse DCT

 ∈ ℝᴴˣᴰ

Denormalize

motor commands

Quantization is lossy; BPE encoding and decoding are lossless with respect to the rounded coefficient sequence.

FAST in π₀.₅

FAST teaches the VLM; the action expert controls the robot

The same demonstrated action chunk has a discrete representation for next-token learning and a continuous representation for flow matching.

Ground-truth continuous action chunk

Pretraining · 280k steps

FAST(A) → VLM

The PaliGemma-based VLM autoregressively predicts FAST action tokens with cross-entropy alongside web, localization, and semantic-subtask data. The flow-matching weight is α = 0, so this stage makes the VLM robot-action-aware before an action expert is trained.

L = −Σᵢ log p(Tᵢ | o, ℓ, T<ᵢ)

Post-training · 80k steps

Noisy Aᵗ → action expert

A randomly initialized ~300M-parameter expert learns a continuous velocity field with flow matching while token prediction continues. The combined loss uses a flow weight of α = 10.

L = L_token + αL_flow

The action expert does not receive predicted FAST tokens. It reads the VLM prefix built from images, language, and robot state, plus a noisy continuous action chunk and flow timestep. Its mask blocks access to FAST action tokens, preventing ground-truth action leakage through the discrete branch.

where the tokens go

π₀-FAST and π₀.₅ use FAST differently

FAST is an inference representation in π₀-FAST, but primarily a pretraining target in π₀.₅.

π₀-FAST

Generate FAST tokens at inference

observationsT₁ … Tₙactions

The autoregressive policy predicts and decodes the compact token sequence into a continuous action chunk.

π₀.₅

Generate continuous actions at inference

VLM prefixactions

The VLM first predicts a textual subtask; the flow-matching expert then produces the low-level continuous chunk in ten integration steps. FAST action tokens are not generated for real-time low-level control, avoiding the roughly 30–60 autoregressive action-token steps used by π₀-FAST.

The mental model

FAST does not assign one token to every scalar action. It first rewrites a complete action chunk in frequency space, where smooth robot motion is concentrated in a few low-frequency coefficients. It rounds those coefficients, orders the broad trajectory before its fine corrections, and uses BPE to merge recurring patterns into a short sequence of discrete tokens.

Think of the result as a motion sentence: early values describe the overall movement across every action dimension, while later values refine it. The only fitted part of the tokenizer is its BPE vocabulary; normalization, DCT, rounding, flattening, and inverse DCT are fixed operations.

Sources

Built by Suveen.

www.suveenellawela.com v.2026.5