Assessing the Impact of Prompting Methods on ChatGPT's Mathematical Capabilities

arXiv:2312.15006v2 Announce Type: replace-cross
Abstract: This study critically evaluates the efficacy of prompting methods in enhancing the mathematical reasoning capability of large language models (LLMs). The investigation uses three prescriptive prompting methods - simple, persona, and conversational prompting - known for their effectiveness in enhancing the linguistic tasks of LLMs. We conduct this analysis on OpenAI's LLM chatbot, ChatGPT-3.5, on extensive problem sets from the MATH, GSM8K, and MMLU datasets, encompassing a broad spectrum of mathematical challenges. A grading script …

abstract arxiv capabilities capability chatgpt conversational cs.ai cs.cl cs.lg impact investigation language language models large language large language models llms mathematical reasoning prompting reasoning simple study tasks type