Prompting & Inference
Chain-of-Thought
Prompting a model to reason step by step, which tends to improve its accuracy on hard problems.
Chain-of-thought prompting asks a model to work through its reasoning out loud before giving a final answer, rather than jumping straight to a conclusion. Laying out intermediate steps markedly improves performance on math, logic, and multi-step questions. Many newer models are trained to do this kind of extended reasoning automatically, spending more effort on harder problems.
Related terms
Prompt Engineering
The practice of wording and structuring prompts to get better, more reliable results from a model.
In-Context Learning
A model’s ability to learn a task from examples placed in the prompt, without any change to its weights.
Prompt
The input you give a generative model, usually instructions or a question, that tells it what you want.