Tree-of-Thought (TOT) 프롬프트 (Tree-of-Thought) 프롬프트 (Tree-of-Thought) 프롬프트 (Tree-of-Thought) 프레임 워크 (Tree-of-Thoughts Framework)에서 아이디어를 빌려주는 새로운 기술을 차용하고 잘 확립 된 생각이 많은 프롬프트 개념을 넓히고 향상시켜 ChatGPT와 같은 대형 언어 모델이 뛰어난 추론 능력을 보여줄 수있게합니다. 이 생각에 대한 추진력이있는 기술은 대형 언어 모델이 지식을 점차적으로 축적하면서 오류를 자율적으로 교정 할 수 있도록합니다.
한 예에서, TOT 프롬프트는 ChatGpt 3.5의 추론 능력을 향상시킵니다.
대형 언어 모델 (LLM)은 종종 특정 유형의 복잡한 질문에 대답하기 위해 고군분투합니다. 대표적인 예는 다음과 같습니다.
Bob is in the living room.
He walks to the kitchen, carrying a cup.
He puts a ball in the cup and carries the cup to the bedroom.
He turns the cup upside down, then walks to the garden.
He puts the cup down in the garden, then walks to the garage.
Where is the ball?
정답은 공이 침실에 있다는 것입니다. LLM은 비 결정적이지만, 여러 번의 시도 중에 내가받은 응답은 다음과 같습니다.
"고려 된 프롬프트 체인"(짧은 COT)으로 알려진 널리 알려진 개념이 존재하며, 이는 큰 언어 모델 (LLM)이 사고 과정을 설명하도록 권장하여 올바른 응답의 가능성을 높입니다. 이 개념의 뿌리는 2022 년 1 월부터 논문으로 거슬러 올라갈 수 있습니다.
핵심으로, COT 기술은 다음과 비슷한 것을 프롬프트에 추가하여 구현할 수 있습니다.
Think carefully and logically, explaining your answer.
이 접근법은 효과가 다르지만 많은 질문에 답할 때 LLM의 성능을 향상시킵니다.
다음은 ChatGpt 3.5의 응답의 예입니다.
Based on the given sequence of events, we can logically determine the location of the ball. Let's analyze each step:
1. Bob is in the living room.
- No mention of the ball in the living room.
2. Bob walks to the kitchen, carrying a cup.
- No mention of the ball in the kitchen yet.
3. Bob puts a ball in the cup.
- The ball is now inside the cup.
4. Bob carries the cup to the bedroom.
- The ball is still inside the cup.
5. Bob turns the cup upside down.
- If the ball was inside the cup, it would have fallen out when Bob turned it upside down.
6. Bob walks to the garden.
- Bob left the cup in the garden, but there is no mention of the ball being removed from the cup.
7. Bob puts the cup down in the garden.
- The ball, if it was in the cup, would still be in the cup in the garden.
8. Bob walks to the garage.
- No mention of the ball being taken to the garage.
Based on the given information, we can conclude that the ball is still inside the cup in the garden.
Chatgpt가 5 단계에서 공이 떨어 졌다는 것을 깨달았지만 여전히 답을 얻지 못했습니다.
Chatgpt 4에 의지하지 않고 더 잘 할 수 있습니까?
최근 몇 주 동안 여러 논문과 파이썬 리포지토리가 등장하여 The Tree-of-thoughts (TOT)의 개념을 소개했습니다.
근본적인 원칙은 LLM에 여러 추론의 분기를 탐색하면서 자체 평가하는 동안 자체 평가하는 기회를 제공하는 것입니다. 참조 된 논문에 따르면,이 접근법은 특정 문제에 매우 효과적인 것으로 보입니다.
THecute Tree Framework에 대해 논의하는 기존 논문은 일반적으로 Python을 통해 LLM에 여러 번 전화를 걸기 위해 전제되는 것으로 보입니다. 추론 체인의 주요 장점 중 하나는 단일 문장 프롬프트로 Chatgpt를 사용하여 달성 할 수 있다는 것입니다.
이로 인해 나는 숙고하게되었습니다. 우리는 Tot 스타일의 프롬프트를 사용하여 COT의 결과를 향상시킬 수 있습니까?
내가 사용한 프롬프트는 다음과 같습니다.
Imagine three different experts are answering this question.
All experts will write down 1 step of their thinking,
then share it with the group.
Then all experts will go on to the next step, etc.
If any expert realises they're wrong at any point then they leave.
The question is...
그리고 다음은 ChatGTP 3.5의 응답입니다.
Expert 1: The ball is in the living room.
Expert 2: The ball is in the kitchen.
Expert 3: The ball is in the bedroom.
Expert 1: Bob carries the cup to the bedroom, so the ball must be in the cup.
Expert 2: Oh, I see my mistake. Yes, the ball is in the cup.
Expert 3: Agreed, the ball is in the cup in the bedroom.
Expert 1: Next, Bob turns the cup upside down in the bedroom.
Expert 2: After that, Bob walks to the garden and puts the cup down.
Expert 3: Therefore, the ball must have fallen out of the cup when Bob turned it upside down in the bedroom. So, the ball is in the bedroom, not in the cup anymore.
Expert 1: Oh, you're right. I made a mistake. The ball is in the bedroom, not in the cup.
Expert 2: Agreed, the ball is in the bedroom.
Expert 3: Bob then walks to the garage, so the ball remains in the bedroom. It is not in the garden or the garage.
Expert 1: Absolutely, the ball is still in the bedroom.
Expert 2: Yes, the ball hasn't moved from the bedroom.
Expert 3: Therefore, the ball is in the bedroom, not in the garden or the garage.
All three experts agree that the ball is in the bedroom.
이번에는 Chatgpt 3.5가 어떻게 답을 얻는 지 알 수 있습니다!
성능 향상을위한 프롬프트를 개선 할 수있는 여지가있어 잠재적으로 더 간결한 응답으로 이어질 수 있습니다.
다른 프롬프트와 함께 성공을 찾으면 PR을 제출하여 tree-of-thought-prompts.txt 에 추가 할 수 있습니다.
고려한 프레임 워크는 여러 이해 관계자와 관련된 조직의 의사 결정 프로세스 와 유사합니다. 이 구조는 합의와 다양한 사고를 모방하려고합니다.
작업의 다른 부분을 처리하도록 미세 조정 된 여러 에이전트를 사용하여 효과를 확장 할 수 있습니다. Autogpt는 이에 대한 좋은 예입니다. 여기서 LLM 계획을 목표로 자체 평가합니다. 전문화 또는 완전히 다른 LLM을 위해 다르게 조정 된 LLM을 사용하면 더 큰 다양한 사고를 허용합니다. 이 LLM은 다른 모드에서 상호 작용할 수 있습니다 (Team Topologies 용어 사용) :
고성능 팀은 종종 의사 결정에서 개인을 능가합니다. 따라서 다른 조직 구조와 특성을 채택하면 LLM의 성능을 향상시킬 수 있습니다. 다양한 사고, 전문화 및 컨벤즈 외에도 모방 할 수 있습니다.
코드를 사용하는 경우이 저장소를 인용하십시오.
@misc{tree-of-thought-prompting,
title = {Using Tree-of-Thought Prompting to boost ChatGPT's reasoning},
author = {Dave Hulbert},
year = 2023,
month = may,
journal = {GitHub repository},
publisher = {Zenodo},
doi = {10.5281/ZENODO.10323452},
url = {https://doi.org/10.5281/zenodo.10323452},
howpublished = {url{https://github.com/dave1010/tree-of-thought-prompting}}
}