Search papers, labs, and topics across Lattice.
2
0
5
1
Agentic AI systems are still far from maximizing hardware potential: SOL-ExecBench reveals a significant gap between current GPU kernel performance and analytically derived Speed-of-Light bounds across a wide range of AI models.
Stabilizing RL training is now possible by modulating importance ratios with hazard-aware penalties, preventing both mode collapse and policy erosion.