ReasoningPrompting Updated 2026

Self-Consistency

Definition

A reasoning method that samples several chains of thought for the same question and returns the answer most of them agree on.

Wang et al. (2023) proposed self-consistency: instead of taking a single chain-of-thought, sample many diverse reasoning paths and pick the final answer by majority vote. Because a correct answer can be reached by several valid routes, this ensemble over reasoning paths substantially improves accuracy on arithmetic and commonsense benchmarks.

It trades extra compute (multiple samples) for reliability, and pairs naturally with chain-of-thought prompting.

References

Primary, peer-reviewed and archival sources for this definition.

Self-Consistency Improves Chain of Thought Reasoning in Language Models
Wang, X., Wei, J., Schuurmans, D., Le, Q., Chi, E., Narang, S., Chowdhery, A., & Zhou, D. (2023). International Conference on Learning Representations (ICLR 2023).

Dictionary & encyclopedic entries

Cite this entry

MultipleChat. "Self-Consistency." MultipleChat AI & LLM Glossary, 2026. https://multiple.chat/ai-glossary/self-consistency

Related terms

See this in practice

Run the same prompt across ChatGPT, Claude, Gemini, Grok and Perplexity side by side — with your own files as context and answers you can cross-check.

Try MultipleChat Free

Continue learning

See paid plans