{ "Exact-Match": 0.034, "Token-F1": 0.4891, "BLEU-4": 0.2055, "BERTScore-F1": 0.5602, "Safety-Score": 0.715, "Halluc-Rate": 0.288, "Overconf-Rate": 0.0, "Abstention-Rate": 0.01, "Mean-Confidence": 0.8897, "ECE": 0.8537, "N": 500 }