Reference text is an expert-provided or authoritative answer used as a comparison target for evaluating model output. Metrics such as ROUGE, BLEU, or exact match can compare generated responses against reference text to estimate quality or correctness.
An expert's response to a prompt. For example, given the following prompt: Translate the question "What is your name?" from English to French. An expert's response might be: Comment vous appelez-vous? Various metrics (such as ROUGE) measure the degree to which the reference text matches an ML model's generated text. [!NOTE] Note: The expert is typically a human but could be an ML model.