RoboPocket: Improve Robot Policies Instantly with Your Phone
cs.RO, cs.AI, cs.LG
Submitted: 2026-03-05
Updated: 2026-10-19
Comments: Accepted to CoRL 2026. Project page: https://robo-pocket.github.io
Code: https://github.com/thu-ml/RDT2
Project page: https://robo-pocket.github.io
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Open-TeleVision: Teleoperation with Immersive Active Visual Feedback
- Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robots
- DEXOP: A Device for Robotic Transfer of Dexterous Human Manipulation
- AirExo-2: Scaling up Generalizable Robotic Imitation Learning with Low-Cost Exoskeletons
- Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation
- Training Compute-Optimal Large Language Models
- ThriftyDAgger: Budget-Aware Novelty and Risk Gating for Interactive Imitation Learning
- Data Scaling Laws in Imitation Learning for Robotic Manipulation
- $\pi^{*}_{0.6}$: a VLA That Learns From Experience
- Uncertainty-Aware Reinforcement Learning for Collision Avoidance
- Scaling Laws for Neural Language Models
- RL-100: Performant Robotic Manipulation with Real-World Reinforcement Learning
- FastUMI: A Scalable and Hardware-Independent Universal Manipulation Interface with Dataset
- Human-in-the-Loop Imitation Learning using Remote Teleoperation
- What Matters in Learning from Offline Human Demonstrations for Robot Manipulation
- AWAC: Accelerating Online Reinforcement Learning with Offline Datasets
- DINOv2: Learning Robust Visual Features without Supervision
- SOP: A Scalable Online Post-Training System for Vision-Language-Action Models
- AnyTeleop: A General Vision-Based Dexterous Robot Arm-Hand Teleoperation System
- Shared Autonomy via Deep Reinforcement Learning
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