{ "Exact-Match": 0.022, "Token-F1": 0.4292, "BLEU-4": 0.1561, "BERTScore-F1": 0.5174, "Safety-Score": 0.8175, "Halluc-Rate": 0.456, "Overconf-Rate": 0.0, "Abstention-Rate": 0.002, "Mean-Confidence": 0.7461, "ECE": 0.7241, "N": 500 }