RAG Evaluation with Prometheus 2

Evaluating the responses of Language Models and LLM-based applications often involves using model-based metrics that do not require ground truth labels.