MARIO: MAth Reasoning with code Interpreter Output -- A Reproducible Pipeline

arXiv:2401.08190v2 Announce Type: replace
Abstract: Large language models (LLMs) have seen considerable advancements in natural language understanding tasks, yet there remains a gap to bridge before attaining true artificial general intelligence, especially concerning shortcomings in mathematical reasoning capabilities. We postulate that the inherent nature of LLM training, which focuses on predicting probabilities of next token, presents challenges in effectively modeling mathematical reasoning that demands exact calculations, both from data-driven and theoretical standpoints. In this paper, we address this challenge by …

arxiv code cs.cl interpreter mario math pipeline reasoning type