🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

🚀Can we have a freely moving hand in the air for the manipulation policy to directly command in the real world? We introduce Flying Hand: End-Effector-Centric Framework for Versatile Aerial Manipulation Teleoperation and Policy Learning. 🎯EE-centric MPC for aerial manipulator

We got a robot to clean up homes that were never seen in its training data! Our new model, π-0.5, aims to tackle open-world generalization. We took our robot into homes that were not in the training data and asked it to clean kitchens and bedrooms. More below⤵️

Meet 𝐀𝐌𝐎 — our universal whole‑body controller that unleashes the 𝐟𝐮𝐥𝐥  kinematic workspace of humanoid robots to the physical world. AMO is a single policy trained with RL + Hybrid Mocap & Trajectory‑Opt. Accepted to #RSS2025. Try our open models & more 👉

🤖Introducing TWIST: Teleoperated Whole-Body Imitation System. We develop a humanoid teleoperation system to enable coordinated, versatile, whole-body movements, using a single neural network. This is our first step toward general-purpose robots. 🌐humanoid-teleop.github.io

our new system trains humanoid robots using data from cell phone videos, enabling skills such as climbing stairs and sitting on chairs in a single policy (w/ Hongsuk Benjamin Choi Junyi Zhang David McAllister)

Impressive work! Yuanhang Zhang

A portable data collection system for general dexterous manipulation—DexWild🦾—can collect data Anywhere, enabling impressive generalization across robots/tasks/scenarios/... Huge congrats to Tony Tao and the team! 💥👏 Website: dexwild.github.io

Tired of teleoperating your robots? We built a way to scale robot datasets without teleop, dynamic simulation, or even robot hardware. Just one smartphone scan + one human hand demo video → thousands of diverse robot trajectories. Trainable by diffusion policy and VLA models

✈️OMW to #ICRA2025 Hope to meet old&new friends and happy to chat about humanoid robot and dex manipulation!😊

What if robots could dream inside a video generative model? Introducing DreamGen, a new engine that scales up robot learning not with fleets of human operators, but with digital dreams in pixels. DreamGen produces massive volumes of neural trajectories - photorealistic robot

Congrats Haoru Xue Chaoyi Pan

Tired of watching fancy humanoid dancing? Can they just do some daily useful tasks like: "Pass me a bottle of Water🍺"? 🤔Turns out it's nontrivial to stablize whole-body manipulation and locomotion at the same time. We basically want our humanoid to be stable as a camera

Bored of linear recurrent memories (e.g., linear attention) and want a scalable, nonlinear alternative? Our new paper “Test-Time Training Done Right” propose LaCT (Large Chunk Test-Time Training) — a highly efficient, massively scalable nonlinear memory with: 💡 Pure PyTorch

Today, we are introducing SmolVLA: a 450M open-source vision-language action model. Best-in-class performance and inference speed! And the best part? We trained it using all the open-source LeRobot datasets in the Hugging Face hub! But how? 🫳🏀

"I believe finding such a scalable off-policy RL algorithm is the most important missing piece in machine learning today." Very insightful blog on offlineRL by Seohong Park 🫡 It's quite painful that offlineRL only works for "reduced horizon" at this stage. looking forward to

💡Wow—super dynamic motion controlled by a unified general policy! 🔗 gmt-humanoid.github.io Feels like the recipe for training a general whole-body controller has almost converged: MoE oracle teacher → generalist student policy In our previous research: - HOVER

VLA language following capability can be improved by freezing the VLM, and a good VLM matters.

Your bimanual manipulators might need a Robot Neck 🤖🦒 Introducing Vision in Action: Learning Active Perception from Human Demonstrations ViA learns task-specific, active perceptual strategies—such as searching, tracking, and focusing—directly from human demos, enabling robust

🚀 GR00T Dreams code is live! NVIDIA GEAR Lab's open-source solution for robotics data via video world models. Fine-tune on any robot, generate 'dreams', extract actions with IDM, and train visuomotor policies with LeRobot datasets (GR00T N1.5, SmolVLA). github.com/NVIDIA/GR00T-D…