Search papers, labs, and topics across Lattice.
KAIST AI
1
0
Tactic-level supervision from symbolic proof assistants can significantly outperform traditional outcome-only reinforcement learning approaches in theorem proving tasks.
