1 paper
Luca Beurer-Kellner, Mark Niklas Müller, Marc Fischer +1
Many recent prompting strategies for large language models (LLMs) query the model multiple times sequentially -- first to produce intermediate results and then the final answer. Ho…