Search papers, labs, and topics across Lattice.
3
0
5
0
WALL-SS achieves unprecedented long-horizon visual simulation for robots, improving action following and trajectory accuracy while maintaining coherence over extended interactions.
DMuon slashes training time for large models, achieving up to 163x faster optimizer steps while maintaining the benefits of matrix-aware updates.
By redefining action learning around semantic events, WALL-WM achieves unprecedented generalization across tasks and environments, outperforming traditional models.
