AI engineering glossary · Prompt context engineering
What is self-consistency?
Self-consistency: A method that samples several different reasoning paths for the same question at a non-zero randomness setting, then takes the most common final answer by majority vote. It improves reliability on reasoning tasks but multiplies the cost by the number of samples.
Self-consistency: an example
Ask a maths question 5 times; answers are 105, 105, 95, 105, 110, so 105 is chosen.
Related terms
More Prompt context engineering terms
- Self-refine
- Skeleton-of-Thought
- Sliding window history
- Stateless API
- Step-back prompting
- Stop sequence
- Structured extraction
- Structured output
Where you learn it
Self-consistency is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.