Which prompt engineering technique is used to encourage large language models to explain their reasoning by revealing the intermediate steps the model takes to arrive at an answer?

Which prompt engineering technique is used to encourage large language models to explain their reasoning by revealing the intermediate steps the model takes to arrive at an answer?

• A. Chain-of-thought (CoT)
• B. Cognitive verifier pattern
• C. Zero-shot
• D. Few-shot

Answer: (A)

The technique aims to elicit step-by-step reasoning from the model, making it reveal the intermediate thinking it uses to reach an answer. This is known as chain-of-thought prompting. By asking the model to think through problems step by step (often with examples that show the reasoning path), the model tends to produce clearer, decomposed solutions for multi-step tasks like math or logic. Zero-shot and few-shot describe how many examples are given, but they don’t inherently require the model to lay out its internal reasoning unless the prompt explicitly asks for it. The cognitive verifier pattern isn’t the standard method for prompting internal reasoning, and it’s not primarily about eliciting chain-of-thought.

The technique aims to elicit step-by-step reasoning from the model, making it reveal the intermediate thinking it uses to reach an answer. This is known as chain-of-thought prompting. By asking the model to think through problems step by step (often with examples that show the reasoning path), the model tends to produce clearer, decomposed solutions for multi-step tasks like math or logic. Zero-shot and few-shot describe how many examples are given, but they don’t inherently require the model to lay out its internal reasoning unless the prompt explicitly asks for it. The cognitive verifier pattern isn’t the standard method for prompting internal reasoning, and it’s not primarily about eliciting chain-of-thought.