Search papers, labs, and topics across Lattice.
1
0
3
Chain-of-Thought prompting doesn't always improve LLMs' ability to solve discrete optimization problems, and surprisingly, "disordered" datasets can sometimes boost performance on simpler tasks.
