What Matters for Latent Actions in Robot Learning
cs.RO, cs.CV
Submitted: 2026-08-20
Updated: 2026-09-26
Comments: Project page: https://carldegio.github.io/latent_action.github.io
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- X-Tokenizer: A Multimodal Action Tokenizer for Vision-Language-Action Pretraining
- FAST: Efficient Action Tokenization for Vision-Language-Action Models
- ActionCodec: What Makes for Good Action Tokenizers
- $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
- OpenVLA: An Open-Source Vision-Language-Action Model
- Qwen3-VL Technical Report
- PaliGemma: A versatile 3B VLM for transfer
- World Action Models are Zero-shot Policies
- Cosmos 3: Omnimodal World Models for Physical AI
- Wan: Open and Advanced Large-Scale Video Generative Models
- World Simulation with Video Foundation Models for Physical AI
- Open-Sora: Democratizing Efficient Video Production for All
- RoboCOIN: An Open-Sourced Bimanual Robotic Data Collection for Integrated Manipulation
- Scalable Vision-Language-Action Model Pretraining for Robotic Manipulation with Real-Life Human Activity Videos
- Emergence of Human to Robot Transfer in Vision-Language-Action Models
- EgoVLA: Learning Vision-Language-Action Models from Egocentric Human Videos
- FastUMI: A Scalable and Hardware-Independent Universal Manipulation Interface with Dataset
- YUBI: Yielding Universal Bidigital Interface for Bimanual Dexterous Manipulation at Scale
- StaMo: Unsupervised Learning of Generalizable Robot Motion from Compact State Representation
- UniVLA: Learning to Act Anywhere with Task-centric Latent Actions
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