Chain-of-Thought (CoT) has become a standard prompting technique for enhancing model performance on complex tasks. The model is instructed to “think step by step” to utilize more test-time computation to decomposing hard tasks into smaller steps.
For GPT3 (2022), users need to give prompts with details step-by-step thinkings for few-shot learning. While for GPT4 (2023), users only have to add “step-by-step” instruction in prompt for same performace.