BLEU and ROUGE are categorized as which type of metrics?

• A. Fairness metrics
• B. Similarity metrics
• C. Accuracy metrics
• D. Parental metrics

Answer: (B)

These metrics measure how similar two pieces of text are, typically a model-generated output and a human reference. BLEU focuses on precision of overlapping n-grams, checking how much of the candidate text’s word sequences appear in the reference and awarding higher scores when there are more exact matches. ROUGE, often used for evaluating summaries, emphasizes recall, assessing how much of the reference content is captured by the candidate through overlapping n-grams and sequences. Because both aim to quantify resemblance between texts rather than classify or judge fairness, they fall squarely under similarity metrics. They’re not about accuracy in a labeling sense, nor about fairness or any unrelated notion like parental metrics.