Assessing the Impact of Prompting Methods on ChatGPT's Mathematical Capabilities
This work addresses the problem of applying prompting strategies to enhance mathematical capabilities in LLMs for researchers and practitioners, showing it is incremental as it tests known methods in a new domain with negative results.
The study evaluated the impact of prompting methods on ChatGPT-3.5's mathematical reasoning using datasets like MATH, GSM8K, and MMLU, finding that none of the methods consistently improved performance and some caused significant degradation.
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 adapted to each dataset is used to determine the effectiveness of these prompting interventions in enhancing the model's mathematical analysis power. Contrary to expectations, our empirical analysis reveals that none of the investigated methods consistently improves over ChatGPT-3.5's baseline performance, with some causing significant degradation. Our findings suggest that prompting strategies do not necessarily generalize to new domains, in this study failing to enhance mathematical performance.