Search papers, labs, and topics across Lattice.
1
0
2
4
LLMs reason better when their uncertainty consistently decreases, paving the way for shorter, more accurate chain-of-thought reasoning.
