Search papers, labs, and topics across Lattice.
Affiliation:
1
0
Large language models can achieve a 28.7-point boost in reasoning accuracy by breaking down complex answer options into simpler atomic judgments.
