ReAT: Residual Action Tokenization for Autoregressive Robot Policies
cs.RO
Submitted: 2026-10-09
Updated: 2026-10-13
Comments: 18 pages, 6 figures, 6 tables
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- FlexTok: Resampling Images into 1D Token Sequences of Flexible Length
- $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
- RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control
- High Fidelity Neural Audio Compression
- ActionCodec: What Makes for Good Action Tokenizers
- OpenVLA: An Open-Source Vision-Language-Action Model
- Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success
- Autoregressive Image Generation using Residual Quantization
- LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning
- OAT: Ordered Action Tokenization
- Finite Scalar Quantization: VQ-VAE Made Simple
- QueST: Self-Supervised Skill Abstractions for Learning Continuous Control
- GR00T N1: An Open Foundation Model for Generalist Humanoid Robots
- FAST: Efficient Action Tokenization for Vision-Language-Action Models
- $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
- Variable Rate Image Compression with Recurrent Neural Networks
- Full Resolution Image Compression with Recurrent Neural Networks
- VQ-VLA: Improving Vision-Language-Action Models via Scaling Vector-Quantized Action Tokenizers
- HiFlow: Tokenization-Free Scale-Wise Autoregressive Policy Learning via Flow Matching
- SoundStream: An End-to-End Neural Audio Codec
Related papers
- FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
- MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
- RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
- ARC: A Reasoning Recipe for Robot Foundation Models
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies