We present Step-Back Prompting, a simple prompting technique that enables LLMs to do abstractions to derive high-level concepts and first principles from instances containing specific details. Using the concepts and principles to guide reasoning, LLMs significantly improve their abilities in following a correct reasoning path towards the solution. We conduct experiments of Step-Back Prompting with PaLM-2L, GPT-4 and Llama2-70B models, and observe substantial performance gains on various challenging reasoning-intensive tasks including STEM, Knowledge QA, and Multi-Hop Reasoning. For instance, Step-Back Prompting improves PaLM-2L performance on MMLU (Physics and Chemistry) by 7% and 11% respectively, TimeQA by 27%, and MuSiQue by 7%.
@article{arxiv.2310.06117,
title = {Take a Step Back: Evoking Reasoning via Abstraction in Large Language Models},
author = {Huaixiu Steven Zheng and Swaroop Mishra and Xinyun Chen and Heng-Tze Cheng and Ed H. Chi and Quoc V Le and Denny Zhou},
journal= {arXiv preprint arXiv:2310.06117},
year = {2024}
}