ClarifyMT-Bench
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
ClarifyMT-Bench
96.6Precision (P)
12
ClarifyMT-Bench Average 1.0 (full dataset)
77.1Accuracy
10
ClarifyMT-Bench Refusal persona 1.0 (full dataset)
72.9Accuracy
10
ClarifyMT-Bench Factually-Wrong persona 1.0
93.7Accuracy
10
ClarifyMT-Bench Contradictory persona 1.0
97.4Accuracy
10
ClarifyMT-Bench Off-Focus persona 1.0
96.1Accuracy
10
ClarifyMT-Bench Partial-Vague persona 1.0
95.5Accuracy
10
ClarifyMT-Bench Precise persona 1.0
96.6Accuracy
10