ActionPiece: Rethinking Action Tokenization for Autoregressive Vision-Language-Action Models

arXiv:2609.18487 · cs.RO, cs.AI, cs.CL, cs.CV · Submitted 2026-09-16 · Read on arXiv

cs.RO, cs.AI, cs.CL, cs.CV

Submitted: 2026-09-16

Updated: 2026-09-16

Comments: Project Page: https://deepcybo-physai.github.io/ActionPiece/

Project page: https://deepcybo-physai.github.io/ActionPiece

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Quantization — 29× in this paper · explained in 12 episodes
• Regularization — 19× in this paper · explained in 3 episodes
• Tokenization — 16× in this paper · explained in 2 episodes
• Benchmark — 14× in this paper · explained in 2 episodes
• Ablation — 13× in this paper · explained in 1 episode
• Fidelity — 13× in this paper · explained in 2 episodes
• Accuracy — 7× in this paper · explained in 1 episode
• Consistency — 7× in this paper · explained in 3 episodes
• Foundation Model — 7× in this paper · explained in 3 episodes
• Reconstruction Error — 7× in this paper · explained in 2 episodes
• Action Space — 6× in this paper · explained in 1 episode
• Generalization — 5× in this paper · explained in 13 episodes
• Physical Rank Consistency — 5× in this paper
• Representation Learning — 5× in this paper · explained in 2 episodes
• Alignment — 4× in this paper · explained in 1 episode
• Vision-Language-Action (VLA) Models — 3× in this paper · explained in 2 episodes
• Benchmarking — 2× in this paper · explained in 3 episodes
• Fine-Tuning — 2× in this paper · explained in 5 episodes
• Foundation Models — 2× in this paper · explained in 7 episodes
• In Annual Conference on Robot Learning — 2× in this paper
• Transformer — 2× in this paper · explained in 2 episodes
• AI Agents — 1× in this paper · explained in 2 episodes
• Action Flow — 1× in this paper · explained in 1 episode
• Catastrophic Forgetting — 1× in this paper · explained in 5 episodes
• Decode — 1× in this paper · explained in 2 episodes
• Exponential Moving Average (EMA) — 1× in this paper · explained in 2 episodes
• Grounding — 1× in this paper · explained in 3 episodes
• Robustness — 1× in this paper · explained in 5 episodes
• Self-Supervised Learning — 1× in this paper · explained in 8 episodes
• Trace — 1× in this paper · explained in 2 episodes
• World Model — 1× in this paper · explained in 7 episodes

Sources

• LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristics
• Motus: A Unified Latent Action World Model
• $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
• Fast-in-Slow: A Dual-System Foundation Model Unifying Fast Manipulation within Slow Reasoning
• StarVLA: A Lego-like Codebase for Vision-Language-Action Model Developing
• ActionCodec: What Makes for Good Action Tokenizers
• Knowledge Insulating Vision-Language-Action Models: Train Fast, Run Fast, Generalize Better
• LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Models
• Categorical Reparameterization with Gumbel-Softmax
• Evo-Depth: A Lightweight Depth-Enhanced Vision-Language-Action Model
• LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning
• Ordered Action Tokens for Visuomotor Policy Learning
• HybridVLA: Collaborative Diffusion and Autoregression in a Unified Vision-Language-Action Model
• FASTer: Toward Efficient Autoregressive Vision Language Action Modeling via Neural Action Tokenization

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
• HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
• APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies
• Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving