Provided by

arXiv

arxiv.org
0.0/5
1 views
0 saved
FreeadvancedText Generation
ResearchersAI EngineersDevelopers
Introduces Chain-of-Thought prompting to significantly improve reasoning abilities and multi-step problem solving in large language models.

Overview

This seminal paper presents Chain-of-Thought (CoT) prompting, a simple yet powerful technique that enables large language models (LLMs) to decompose complex, multi-step problems into intermediate reasoning steps. By providing examples of step-by-step thinking, CoT prompting drastically improves LLMs' performance on tasks requiring mathematical, symbolic, and common sense reasoning. The method demonstrates that LLMs can achieve better accuracy by articulating their reasoning process, making it a foundational advancement for enhancing AI's problem-solving capabilities.

Abstract

Large language models (LLMs) have shown remarkable few-shot performance on a range of tasks. We explore how to unlock the reasoning abilities of LLMs via a simple prompting method called chain-of-thought prompting. We show that chain-of-thought prompting enables models to decompose multi-step problems into intermediate steps, which can be useful for solving problems that require reasoning. Experiments on mathematical reasoning (GSM8K, Math23K, AQuA-RAT), symbolic reasoning (Last Letter Concatenation, Coin Flip), and common sense reasoning (CSQA, StrategyQA, Sports Understanding) benchmarks demonstrate that chain-of-thought prompting improves the performance of LLMs on these tasks. For example, chain-of-thought prompting improves the performance of a 137B parameter model on GSM8K from 17.9% to 58.1%, on AQuA-RAT from 29.3% to 47.2%, and on StrategyQA from 59.8% to 75.3%. The improvements are especially noticeable for more complex reasoning tasks that require multiple steps. Our results suggest that chain-of-thought prompting may be a general method for improving the reasoning abilities of LLMs.