Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

CRSArena

Benchmarks

Task NameDataset NameSOTA ResultTrend
Conversation EvaluationCRSArena Eval (All)
Pearson Correlation (r)0.918
9
Conversation EvaluationCRSArena-Eval Dial-level
Pearson r0.913
9
Conversation EvaluationCRSArena-Eval Turn-level
Pearson Correlation (r)0.93
9
System Ranking CorrelationCRSArena-Eval (All)
Pearson Correlation (r)0.918
9
System Ranking CorrelationCRSArena-Eval Dial-level
Pearson Correlation (r)0.913
9
System Ranking CorrelationCRSArena-Eval Turn-level
Pearson Correlation (r)0.93
9
Showing 6 of 6 rows