A Brief Study of Prompting Techniques for Reasoning Tasks

In recent times, there has been a growing interest in the capabilities of large language models (LLMs). Prompting techniques give an efficient way to guide the LLM and get better results. The process of designing a prompt that can generate the desired output is iterative. Here, we review different prompting techniques available for solving complex reasoning tasks. It discusses how these techniques affect the accuracy and efficiency of LLMs when they are used to address reasoning tasks. Furthermore, this survey may work as a starting point for individuals who are new to the field of prompting. This work also contains a comparative analysis of the results of different prompting techniques to assess their accuracy.