Search papers, labs, and topics across Lattice.
6
0
6
4
Achieving a 57.6% success rate on RoboCasa365, Xiaomi-Robotics-1 sets a new standard for vision-language-action models in real-world robotic manipulation.
Xiaomi-Robotics-U0 achieves a remarkable 63.2% success rate on challenging real-world manipulation tasks, showcasing the power of foundation models in embodied robotics.
VAIC enables humanoid robots to perform complex tasks in real-world settings without the need for perfect state observability, significantly advancing their practical deployment.
Achieve state-of-the-art performance in vision-language-action tasks with Xiaomi-Robotics-0, a model that executes smoothly in real-time on real robots using a consumer-grade GPU.
Forget text instructions: bounding-box guidance unlocks surprisingly effective data scaling laws for semantic manipulation in real-world robotics.
Humanoid robots can now perform agile tasks like skateboarding and cart-pushing with underactuated objects, thanks to a dynamics-aware world model that predicts object states from robot's own sensory history.
