思考の連鎖と自己一貫性プロンプティング

OpenAI API によるプロンプトエンジニアリング

Fouad Trad

Machine Learning Engineer

思考の連鎖プロンプト

LLMに、回答を出す前に推論のステップ（思考）を示すよう求める
複雑な推論タスクに使用
モデルのエラーを減らすのに役立つ

Image showing how a chain-of-thought prompt asks the model to solve a problem step by step, and the output contains the reasoning for each step followed.

思考の連鎖プロンプト

推論タスクを解くための標準的なプロンプティング

prompt = """Q: You start with 15 books in your collection. At the bookstore, you 
purchase 8 new books. Then, you lend 3 to your friend and 2 to your cousin. Later, 
you visit another bookstore and buy 5 more books. How many books do you have now? 
A: The answer is"""

print(get_response(prompt))

25 books

思考の連鎖プロンプト

推論タスクを解くための思考の連鎖プロンプティング

prompt = """Q: You start with 15 books in your collection. At the bookstore, you 
purchase 8 new books. Then, you lend 3 to your friend and 2 to your cousin. Later, 
you visit another bookstore and buy 5 more books. How many books do you have now? 
A: Let's think step by step""" 
print(get_response(prompt))

Step 1: Start with the number of books in your collection: 15 books
Step 2: Purchase 8 new books at the bookstore: 15 + 8 = 23 books
Step 3: Lend 3 books to your friend: 23 - 3 = 20 books
Step 4: Lend 2 books to your cousin: 20 - 2 = 18 books
Step 5: Visit another bookstore and buy 5 more books: 18 + 5 = 23 books
Therefore, you have 23 books now.

few-shotを使った思考の連鎖プロンプティング

example = """
Q: The odd numbers in this group add up to an even number:  9, 10, 13, 4, 2.
A: Adding all the odd numbers (9, 13) gives 22. The answer is True.
"""

question = """
Q: The odd numbers in this group add up to an even number: 15, 13, 82, 7. 
A:
"""

prompt = example + question
print(get_response(prompt))

Adding all the odd numbers (15, 13, 7) gives 35. The answer is False.

思考の連鎖とマルチステップ・プロンプティングの比較

マルチステップ・プロンプト

プロンプト内に手順を組み込む

Diagram showing that a multi-step prompt is a prompt that gives the model a series of sequential steps to perform.

思考の連鎖とマルチステップ・プロンプティングの比較

マルチステップ・プロンプト

プロンプト内に手順を組み込む

Diagram showing that a multi-step prompt is a prompt that gives the model a series of sequential steps to perform.

思考の連鎖プロンプト

モデルに中間ステップを生成するよう求める

Image showing that in chain-of-thought prompts, the steps are produced by the model while generating the output.

思考の連鎖の限界

1つの思考がうまくいかない → 結果もうまくいかない
自己一貫性プロンプトが導入されました

Diagram for a chain-of-thought prompt highlighting that one thought with flawed reasoning will lead to an unsuccessful outcome.

自己一貫性プロンプティング

モデルに複数回プロンプトを与えることで、複数の思考の連鎖を生成する
多数決で最終出力を得る

Image showing that a self-consistency prompt is just about multiple chain-of-thoughts, each of which having an output, and the final outcome is obtained by majority vote.

自己一貫性プロンプティング

複数のプロンプトを定義するか、複数の回答を生成するプロンプトを使用して実行できます。

self_consistency_instruction = "Imagine three completely independent experts who 
reason differently are answering this question. The final answer is obtained by 
majority vote. The question is: "


problem_to_solve = "If there are 10 cars in the parking lot and 3 more cars arrive. 
Half the original number of cars leave. Then, half of the current number of cars 
arrive. How many cars are there in the parking?"


prompt = self_consistency_instruction + problem_to_solve

print(get_response(prompt))

自己一貫性プロンプト

Expert 1: Let's go step by step [...] Therefore, the total number of cars in the 
parking lot is 8 + 4 = 12.

Expert 2: First, let's calculate [...] Therefore, the total number of cars in the 
parking lot is now 5 + 2 = 7 cars.

Expert 3: Initially, there are 10 cars [...] Thus, the final answer is 8 + 4 = 12 
cars in the parking lot.

Based on the majority vote, the final answer is that there are 12 cars in the 
parking lot.

練習しましょう！

OpenAI API によるプロンプトエンジニアリング