GPT-3.5, GPT-4, or BARD? Evaluating LLMs Reasoning Ability in Zero-Shot Setting and Performance Boosting Through Prompts
This provides empirical benchmarks for LLM reasoning, aiding researchers and practitioners in model selection and prompt optimization.
The paper evaluated the reasoning abilities of GPT-3.5, GPT-4, and BARD on eleven datasets, finding GPT-4 superior in zero-shot settings, and proposed engineered prompts to boost performance.
Large Language Models (LLMs) have exhibited remarkable performance on various Natural Language Processing (NLP) tasks. However, there is a current hot debate regarding their reasoning capacity. In this paper, we examine the performance of GPT-3.5, GPT-4, and BARD models, by performing a thorough technical evaluation on different reasoning tasks across eleven distinct datasets. Our paper provides empirical evidence showcasing the superior performance of ChatGPT-4 in comparison to both ChatGPT-3.5 and BARD in zero-shot setting throughout almost all evaluated tasks. While the superiority of GPT-4 compared to GPT-3.5 might be explained by its larger size and NLP efficiency, this was not evident for BARD. We also demonstrate that the three models show limited proficiency in Inductive, Mathematical, and Multi-hop Reasoning Tasks. To bolster our findings, we present a detailed and comprehensive analysis of the results from these three models. Furthermore, we propose a set of engineered prompts that enhances the zero-shot setting performance of all three models.