Search papers, labs, and topics across Lattice.
Institute of Automation, Chinese Academy of Sciences
3
0
5
19
FeelWorld achieves a 61% reduction in prediction error compared to visual-only baselines, revolutionizing how we model tactile interactions in robotics.
CoRe achieves preference-aligned reinforcement learning by seamlessly integrating human-like reward decomposition, outperforming traditional methods in both simulated and real-world environments.
Tactile robotic perception gets a boost with a new pretraining method that explicitly encodes force, geometry, and orientation, leading to a 52% reduction in regression error.
