Search papers, labs, and topics across Lattice.
1
0
3
LLM agents can learn to solve tasks previously beyond their reach by exploring high-level language strategies instead of low-level actions, leading to more efficient and effective reinforcement learning.
