Search papers, labs, and topics across Lattice.
3
0
3
2
MoST achieves up to 717% performance improvement by leveraging diverse knowledge sources and cross-scenario strategies for code optimization.
LLMs can now automatically generate verified, project-scale software that passes rigorous testing, thanks to a novel interactive theorem proving approach.
RL agents can learn to write stronger formal specifications by using automatically generated negative tests as a reward signal for completeness, outperforming standard verification-based rewards.
