aiqbench analysis
This paper introduces CHORD, a reinforcement learning framework that leverages human motion-capture data by modeling interactions through object-centric contact wrench spaces to improve scalability in dexterous manipulation tasks. By establishing a large-scale simulation benchmark with nearly 5,000 bimanual tasks, the study demonstrates significant success rates and effective transfer of learned policies from simulation to real-world robotic systems. The approach advances machine learning techniques for complex robot control without relying on quantum computing or consumer electronics applications.
Learning Dexterous Manipulation Using Contact Wrench Guidance From Human Demonstration
arXiv:2607.00033v2 Announce Type: replace Abstract: Dexterous robot manipulation can benefit from the abundance of human demonstrations, but transferring such demonstrations to robot policies remains challenging. We present Contact Wrench Guidance from Human Demonstration in Robotic Dexterous Manipulation (CHORD), a framework for long-horizon manipulation of rigid and articulated objects with reinforcement learning. The key idea is object-centric contact wrench space guidance: we represent human and robot motions by the forces and torques they can induce on the object, enabling similarity to be measured by the induced instantaneous motions. This guidance makes reinforcement learning more scalable for contact-rich dexterous manipulation. We further introduce a large-scale simulation benchmark with 4,739 bimanual dexterous manipulation tasks, constructed from motion-capture datasets and reconstructed in-house videos. Evaluated on 1,831 benchmark tasks, CHORD achieves an average success rate of 82.12%, demonstrating strong scalability. CHORD also generalizes to whole-body manipulation from hand-only and third-person demonstrations, achieving a 90.77% success rate, and the learned policies transfer to the real world in both open-loop and closed-loop settings.
Abstract:Dexterous robot manipulation can benefit from the abundance of human demonstrations, but transferring such demonstrations to robot policies remains challenging. We present Contact Wrench Guidance from Human Demonstration in Robotic Dexterous Manipulation (CHORD), a framework for long-horizon manipulation of rigid and articulated objects with reinforcement learning. The key idea is object-centric contact wrench space guidance: we represent human and robot motions by the forces and torques they can induce on the object, enabling similarity to be measured by the induced instantaneous motions. This guidance makes reinforcement learning more scalable for contact-rich dexterous manipulation. We further introduce a large-scale simulation benchmark with 4,739 bimanual dexterous manipulation tasks, constructed from motion-capture datasets and reconstructed in-house videos. Evaluated on 1,831 benchmark tasks, CHORD achieves an average success rate of 82.12%, demonstrating strong scalability. CHORD also generalizes to whole-body manipulation from hand-only and third-person demonstrations, achieving a 90.77% success rate, and the learned policies transfer to the real world in both open-loop and closed-loop settings.
Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
arXiv:2607.00033 [cs.RO]
(or arXiv:2607.00033v2 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2607.00033 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 261 words
Related coverage
The State of Simulation for Physical AI: An Overview
Hugging Face · Jul 21, 2026
DexTaG: Tactile-as-Guidance in Reinforcement Learning for Dexterous Manipulation
arXiv · Sep 29, 2026
DexMani: Human-Derived Manipulability Guidance for Dexterous Rotation
arXiv · Aug 5, 2026
FastGrasp: Learning-based Whole-Body Control Method for Fast Dexterous Grasping with Mobile Manipulators
arXiv · Sep 29, 2026