1 paper
Junmo Cho, Suhan Kim, Sangjune An +5
Finding effective prompts for language models (LMs) is critical yet notoriously difficult: the prompt space is combinatorially large, rewards are sparse due to expensive target-LM…