Self-Consistency Improves Chain of Thought Reasoning in Language Models

by Xuezhi Wang et al.

Sample several reasoning paths and take the majority answer.

This research explores choosing an answer by comparing multiple generated reasoning attempts. It highlights a quality-versus-computation trade-off that should be evaluated on the intended task.

Level: Advanced
Models: any
Category: Prompting
Tags: reasoning, research
Source date: 2022-03-21
Added: 2026-10-18
Last verified: 2026-10-18
Status: ok

Summary in our own words. All rights with the original creator. Correction or removal: DM on X.