Search papers, labs, and topics across Lattice.
3
0
5
A physics-informed neural network can dramatically improve tension prediction in parachute deployment, outperforming traditional methods in both speed and accuracy.
LLMs can now resolve knowledge conflicts in real-time, leading to more reliable and interpretable outputs.
Robot RL training can be dramatically sped up (3-10x) by decoupling CPU-based simulation from GPU-based learning, challenging the assumption that GPU-resident physics is essential for efficiency.
