We’ve long faced the gap between strong power grasps and fine precision grasps — high-dof hands rarely do both well. So we turned to co-design: optimizing control and fingertip geometry together. The result is a single hand that achieves both, almost like having a gripper and a

A large human behavior model. Introducing In-N-On, our latest findings in scaling egocentric data for humanoids. 1. Pre-training and post-training with human data 2. 1,000+ hours of in-the-wild data and 20+ hours of on-task data with accurate action labels Website:

Meet ACE-F — a novel, foldable teleoperation platform for collecting high-quality robot demonstration data across robot embodiments. Using a specialized soft-controller pipeline, we interpret end-effector positional deviations as virtual force signals to provide the user with

We align sim-real with contacts!

For years, researchers have believed that simulation could help scale up data, but physics remains the main bottleneck for sim-to-real deployment. Aligning simulation with real-world physics using visual 👁️ and contact 🤝 observations through a network could be a promising step.

A neural dynamics model learned in simulation requires calibration with real-world multimodal data to bridge the physics gap.

A lot of people ask me about the sim2real physics gap. I would say that it’s also highly important and at the same time non-trivial to bridge. Here’s a solution but obviously there’s a long way to go.

Very exciting work!🤩🤩🤩

Learning world models from touch instead of pixels. Led by Changwei Jing and yisha

interesting. We also observed that contact predictions help locomotion control. I wonder if this would be general for learning-based control if we first predict the contacts, e.g., for VLA/WMs.

World dynamics models should capture and condition on granular contact information

"Cross-embodiment" is a sign of generalization. We’ve seen huge progress in manipulation and navigation — but what about humanoid whole-body control? Can ONE policy control multiple different humanoids? Meet our #ICRA2026 work 🦅EAGLE: Embodiment-Aware Generalist Specialist

Saining Xie Perfect timing! Cool work on learning world models from touch! x.com/i/status/20191…