What is a side-by-side evaluation?
Comparing the quality of two models by judging their responses to the same prompt.
side-by-side evaluation explained in plain English
Comparing the quality of two models by judging their responses to the same prompt. For example, suppose the following prompt is given to two different models: Create an image of a cute dog juggling three balls. In a side-by-side evaluation, a rater would pick which image was "better" (More accurate? More beautiful? Cuter?).
Example
Practitioners refer to side-by-side evaluation when building, training, or evaluating machine learning systems. It appears in research papers, product documentation, and technical discussions about AI capabilities and limitations.
People also read
- average precision at k
A metric for summarizing a model's performance on a single prompt that generates ranked results, such as a numbered list of book recommendations.
- black box model
A model whose "reasoning" is impossible or difficult for humans to understand.
- Chain-of-Thought Prompting
Asking an AI to show its reasoning step by step before giving a final answer, which often improves accuracy on complex tasks.
- conversational coding
An iterative dialog between you and a generative AI model for the purpose of creating software.
- deterministic
A system that always returns the same output for a given input.
- direct prompting
Synonym for zero-shot prompting.
- distillation
The process of reducing the size of one model (known as the teacher) into a smaller model (known as the student) that emulates the original model's predictions as faithfully as possible.
- emergent behavior
An LLM's ability to generate responses for prompts that it was not explicitly trained on.
- factuality
Within the ML world, a property describing a model whose output is based on reality.
- few-shot learning
A machine learning approach, often used for object classification, designed to train effective classification models from only a small number of training examples.