Chain-of-Thought (CoT) Prompting

대규모 언어 모델(LLMs)이 복잡한 문제를 해결할 때 여러 단계로 나눠서 생각하게 하는 방법입니다.

모델의 추론 능력을 향상시켜 더 복잡한 문제를 해결할 때 도움이 된다고 합니다.

 

[2201.11903] Chain-of-Thought Prompting Elicits Reasoning in Large Language Models (arxiv.org) 의 예시를 보면

 

---

[Standard Prompting]

Model Input:

Q: Roger has 5 tennis balls. He buys 2 more cans of tennis balls. Each can has 3 tennis balls. How many tennis balls does he have now?

A: The answer is 11 .

Q: The cafeteria had 23 apples. If they used 20 to make lunch and bought 6 more, how many apples do they have?

 

Model Output:

A: The answer is 27.

 

답은 9이므로 틀린 답변 입니다.

---

 

[Chain-of-Thought Prompting]

Model Input:

Q: Roger has 5 tennis balls. He buys 2 more cans of tennis balls. Each can has 3 tennis balls. How many tennis balls does he have now?
A: Roger started with 5 balls. 2 cans of 3 tennis balls ach is 6 tennis balls. 5 + 6 = 11. The answer is 11 .

Q: The cafeteria had 23 apples. If they used 20 to make lunch and bought 6 more, how many apples do they have?

 

Model Output:

A: The cafeteria had 23 apples originally. They used 20 to make lunch. So they had 23 - 20 = 3. They bought 6 more apples, so they have 3 + 6 = 9 The answer is 9.

 

이 경우 답을 도출하는 추론단계를 제시함으로 모델이 9로 맞는 답변을 하는 모습을 볼 수 있습니다.

---

 

위 예시는 현제 최신 업데이트 된 ChatGPT 4o에서는 Standard Prompting 으로도 정상적인 답변이 도출되고 있습니다. (여러번 질문 해봄)

 

 

내용 출처:

https://www.ibm.com/topics/chain-of-thoughts

What is Chain of Thoughts (CoT)? | IBM

https://deepgram.com/learn/chain-of-thought-prompting-guide

Chain-of-Thought Prompting: Helping LLMs Learn by Example | Deepgram

https://ar5iv.labs.arxiv.org/html/2201.11903

Chain-of-Thought Prompting Elicits Reasoning in Large Language Models

https://attri.ai/generative-ai-wiki/chain-of-thought-prompting

Chain-of-Thought Prompting | Generative AI Wiki

https://arxiv.org/abs/2201.11903

Chain-of-Thought Prompting Elicits Reasoning in Large Language Models