Evaluation Metrics for GenAI Quiz

What you'll be asked

1. Which evaluation metric is primarily used to measure the degree of overlap between the generated output and a reference ground truth text?
2. When building a RAG (Retrieval-Augmented Generation) pipeline, which metric is most effective for evaluating the 'Faithfulness' of the generated answer to the retrieved context?
3. You are evaluating a chatbot's ability to remain helpful and harmless. Which approach is considered the 'Gold Standard' for high-quality, nuanced evaluation of conversational tone?
4. If you use an LLM-as-a-judge to evaluate your application, what is the primary risk you must mitigate to ensure the evaluation is reliable?
5. In a RAG system, your 'Context Recall' score is low, but your 'Faithfulness' score is high. What does this indicate about your system?

Answers and explanations come after you submit.