Search papers, labs, and topics across Lattice.
1
5
3
9
Continuous scoring from LLM-as-a-Verifier leads to state-of-the-art verification accuracy and improved sample efficiency in reinforcement learning tasks.
