12 datasets
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
12 datasets mean performance (test)
99.57Fidelity
19
12 datasets mean over 32 replicates
0.786Offline AUC
12
12 datasets mean over 32 replicates
25.4Offline TPR @ FPR=0.01
12
Loading the SOTA2 catalog…