English

Take a Step Back: Evoking Reasoning via Abstraction in Large Language Models

Machine Learning 2024-03-13 v2 Artificial Intelligence Computation and Language

Abstract

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%.

Keywords

Cite

@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}
}

Comments

ICLR 2024

R2 v1 2026-06-28T12:45:13.905Z