Search papers, labs, and topics across Lattice.
3
0
5
3
A portable IMU-based teleoperation system slashes the data requirements for humanoid robot manipulation by pre-training on human motion and then fine-tuning on robot data.
Forget scaling laws: this humanoid robot model crushes benchmarks using 10x less data by cleverly pre-training on human videos and then fine-tuning on robot-specific movements.
Achieve more precise and generalizable robot grasping by injecting grasp pose priors into latent diffusion policies, outperforming existing imitation learning methods.
