Beyond Reconstruction: What Matters in Action Tokenization for Robot Policies?
cs.RO
Submitted: 2026-10-06
Updated: 2026-10-06
Terminology
Sources
- RT-1: Robotics Transformer for Real-World Control at Scale
- ActionCodec: What Makes for Good Action Tokenizers
- MaxUp: A Simple Way to Improve Generalization of Neural Network Training
- VLA-0: Building State-of-the-Art VLAs with Zero Modification
- SA-VLA: State-aware tokenizer for improving Vision-Language-Action Models' performance
- X-Tokenizer: A Multimodal Action Tokenizer for Vision-Language-Action Pretraining
- OpenVLA: An Open-Source Vision-Language-Action Model
- Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success
- Behavior Generation with Latent Actions
- SmolVLM: Redefining small and efficient multimodal models
- GR00T N1: An Open Foundation Model for Generalist Humanoid Robots
- $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
- Robust Latent Matters: Boosting Image Generation with Sampling Error Synthesis
- Action Chunking and Exploratory Data Collection Yield Exponential Improvements in Behavior Cloning for Continuous Control
- Double Preconditioning (DoPr): Optimization for Test-Time Performance, not Validation Loss
- Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware
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