1 paper
Jiaxin Cui, Rohan Alexander
We systematically evaluate the reproducibility of data analysis conducted by Large Language Models (LLMs). We evaluate two prompting strategies, six models, and four temperature se…