HyperAI超神经

Dialogue Evaluation On Usr Topicalchat

评估指标

Pearson Correlation
Spearman Correlation

评测结果

各个模型在此基准测试上的表现结果

比较表格
模型名称Pearson CorrelationSpearman Correlation
usr-an-unsupervised-and-reference-free0.42200.4192
proxy-indicators-for-the-quality-of-open0.49740.4877
mdd-eval-self-training-on-augmented-data-for0.45750.5109
usr-an-unsupervised-and-reference-free0.40680.3245
usr-an-unsupervised-and-reference-free0.32210.1419
usr-an-unsupervised-and-reference-free0.33450.3086