ModelRefs / Step-Back Prompting — AI Glossary
Step-Back Prompting — AI Glossary
Prompting the model to first answer a higher-level abstraction of a question before answering the specific question.
Overview
Step-back prompting (Zheng et al. 2023, Google DeepMind) improves multi-hop reasoning: ask 'What physics principles govern this phenomenon?' before 'Why does X happen?' The model retrieves relevant high-level knowledge before applying it. Improves performance on MMLU, TimeQA, and multi-step physics problems.
Reference details
| Topic | prompting |
|---|---|
| Last reviewed | 2026-06-24 |
Related terms
Commonly confused with
Three techniques get confused because all three add a step before the answer. Step-back asks a more general version of the question first. Generated-knowledge asks for relevant facts first. Chain-of-thought works the specific problem in the open. Step-back is the one that helps when the question is too narrow to trigger the right knowledge, rather than too complex to do in one pass.
Primary source
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to Step-Back Prompting — AI Glossary.
Frequently asked questions
What is Step-Back Prompting?
Prompting the model to first answer a higher-level abstraction of a question before answering the specific question.
What concepts are related to Step-Back Prompting?
Closely related concepts include chain of thought, least to most prompting, reasoning.