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

1 Introduction

1. What is chain-of-thought prompting?
   Few-shot prompting where each exemplar is a triple ⟨input, chain of thought, output⟩: the demonstrations include a series of intermediate natural-language reasoning steps before the final answer, so the model generates its own reasoning steps for the test question.

   invented exemplar: “A library has 12 shelves with 8 books each and lends out 30 books. How many books are left?”
   standard	A: The answer is 66.
   chain of thought	A: 12 shelves with 8 books each is 12 × 8 = 96 books. After lending 30, 96 − 30 = 66 are left. The answer is 66.

   • Wei et al., 2022, Chain-of-Thought Prompting Elicits Reasoning in Large Language Models (Advances in Neural Information Processing Systems 35 (NeurIPS 2022)), §1
2. Which two ideas does chain-of-thought prompting combine, and what limits did each have?
   Natural-language rationales help arithmetic reasoning, but training or fine-tuning on rationales needs costly annotated data; few-shot prompting needs no training, but standard prompting does poorly on reasoning and often barely improves with scale. CoT prompting gets rationales through prompting alone.
   • Wei et al., 2022, Chain-of-Thought Prompting Elicits Reasoning in Large Language Models (Advances in Neural Information Processing Systems 35 (NeurIPS 2022)), §1
3. What was the headline result of chain-of-thought prompting?
   PaLM 540B with eight CoT exemplars reached about 57% on GSM8K math word problems, a new state of the art above fine-tuned GPT-3 with a verifier (55%), whereas standard prompting got about 18%.
   • Wei et al., 2022, Chain-of-Thought Prompting Elicits Reasoning in Large Language Models (Advances in Neural Information Processing Systems 35 (NeurIPS 2022)), §1