XRZero-G0: Pushing the Frontier of Dexterous Robotic Manipulation with Interfaces, Quality and Ratios
2026-04-14 • Robotics
Robotics
AI summaryⓘ
The authors developed XRZero-G0, a system that uses virtual reality and special robot grippers to collect high-quality action data without needing a physical robot. Their process improves data quality and efficiency with a feedback loop that checks and trains on the data, achieving 85% validity. They found that mixing a small amount of real robot data with a large amount of robot-free data can perform as well as only using real robot demonstrations, but much cheaper. Using XRZero-G0, they created a large dataset that helps teach real robots to perform tasks without prior training on those robots.
dexterous robot manipulationfoundation modelsvirtual reality interfacerobot-free human demonstrationsdata collection pipelineclosed-loop workflowpolicy learningcross-embodiment transferdata validity ratedataset mixing strategies
Authors
Junming Wang, Teng Pu, Wingmun Fung, Jindong Wang, Shanchang Wang, Yuan Deng, Shuyuan Wang, Ziwei Liu, Kunhao Pan, Ping Yang, Peng Zhai, Yuxin Liang, Xiaofan Li, Jiabi Sun, Renchao Xu, Xiaotian Tian, Pengfei Yan, Guoqiang Ye, Liang Li, Qian Wang, Ruyi Gan, Hao Wang
Abstract
The acquisition of high-quality, action-aligned demonstration data remains a fundamental bottleneck in scaling foundation models for dexterous robot manipulation. Although robot-free human demonstrations (e.g., the UMI paradigm) offer a scalable alternative to traditional teleoperation, current systems are constrained by sub-optimal hardware ergonomics, open-loop workflows, and a lack of systematic data-mixing strategies. To address these limitations, we present XRZero-G0, a hardware-software co-designed system for embodied data collection and policy learning. The system features an ergonomic, virtual reality interface equipped with a top-view camera and dual specialized grippers to directly improve collection efficiency. To ensure dataset reliability, we propose a closed-loop collection, inspection, training, and evaluation pipeline for non-proprioceptive data. This workflow achieves an 85% data validity rate and establishes a transparent mechanism for quality control. Furthermore, we investigate the empirical scaling behaviors and optimal mixing ratios of robot-free data. Extensive experiments indicate that combining a minimal volume of real-robot data with large-scale robot-free data (e.g., a 10:1 ratio) achieves performance comparable to exclusively real-robot datasets, while reducing acquisition costs by a factor of twenty. Utilizing XRZero-G0, we construct a 2,000-hour robot-free dataset that enables zero-shot cross-embodiment transfer to a target physical robot, demonstrating a highly scalable methodology for generalized real-world manipulation.Our project repository: https://github.com/X-Square-Robot/XRZero-G0