aiqbench analysis
ReDex addresses the critical sim-to-real gap in dexterous robotics by enabling a hybrid control scheme that combines frozen simulation policies with human-corrected tactile feedback at specific fingers. This approach successfully integrates force regulation into proprioception-only models, significantly improving success rates on complex manipulation tasks like object flipping and screwdriver rotation without requiring full tactile simulation.
ReDex: Repairing Sim-to-Real Dexterous Policies by Finger-Level Compliant Interaction
arXiv:2610.07525v1 Announce Type: new Abstract: Dexterous manipulation policies trained in simulation often fail to transfer to the real world because of errors in contact timing and force regulation. Yet these policies can retain useful multi-finger coordination for task progression. We propose ReDex, a framework for adapting a simulation-trained base policy to the real world by correcting local contact failures and incorporating tactile feedback. Starting from a proprioception-only base policy, ReDex allows a human operator to physically correct contact failures at selected fingers under compliant control during real-world rollouts, while the frozen base policy continues to control the remaining fingers. These rollouts combine base policy execution, human-corrected finger motion, and fingertip force observations. We reconstruct force-informed targets from these rollouts to train a standalone force-conditioned policy via behavior cloning. This design reduces human correction effort, enables learning of contact regulation from real-world interaction, and introduces force feedback into a proprioception-only policy without tactile simulation or complex full-hand teleoperation. We evaluate ReDex on two challenging, contact-rich dexterous manipulation tasks on real hardware. Compared with sim-to-real transferred base policies, ReDex increases Object Flipping success rate from 14\% to 86\% across two objects and average Screwdriver Rotation progress from 26.0% to 95.3% across three objects.
Abstract:Dexterous manipulation policies trained in simulation often fail to transfer to the real world because of errors in contact timing and force regulation. Yet these policies can retain useful multi-finger coordination for task progression. We propose ReDex, a framework for adapting a simulation-trained base policy to the real world by correcting local contact failures and incorporating tactile feedback. Starting from a proprioception-only base policy, ReDex allows a human operator to physically correct contact failures at selected fingers under compliant control during real-world rollouts, while the frozen base policy continues to control the remaining fingers. These rollouts combine base policy execution, human-corrected finger motion, and fingertip force observations. We reconstruct force-informed targets from these rollouts to train a standalone force-conditioned policy via behavior cloning. This design reduces human correction effort, enables learning of contact regulation from real-world interaction, and introduces force feedback into a proprioception-only policy without tactile simulation or complex full-hand teleoperation. We evaluate ReDex on two challenging, contact-rich dexterous manipulation tasks on real hardware. Compared with sim-to-real transferred base policies, ReDex increases Object Flipping success rate from 14\% to 86\% across two objects and average Screwdriver Rotation progress from 26.0% to 95.3% across three objects.
Robotics (cs.RO)
arXiv:2610.07525 [cs.RO]
(or arXiv:2610.07525v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.07525 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
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 · 289 words
Related coverage
The Machines that Make the Machines
NVIDIA · Oct 7, 2026
Blind Dexterous Grasping via Real2Sim2Real Tactile Policy Learning
arXiv · Jun 12, 2026
Blind Dexterous Grasping via Real2Sim2Real Tactile Policy Learning
arXiv · Jun 11, 2026
Dexterous Point Policy: Learning Point-based Dexterous Hand Policies from Human Demonstrations
arXiv · Jun 10, 2026