HyperAI

Dialogue Evaluation On Usr Personachat

Metrics

Pearson Correlation
Spearman Correlation

Results

Performance results of various models on this benchmark

Comparison Table
Model NamePearson CorrelationSpearman Correlation
proxy-indicators-for-the-quality-of-open0.52900.5382
usr-an-unsupervised-and-reference-free0.07880.0795
usr-an-unsupervised-and-reference-free-0.0454-0.0495
usr-an-unsupervised-and-reference-free0.60870.4814
usr-an-unsupervised-and-reference-free0.41150.4693