PSI
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
PSI
18.91ADE
21
PSI (test)
63.09Accuracy
9
PSI 1.0 (test)
88Accuracy
9
PSI twoCoins (F) (test)
0Inference Time (s)
6
PSI benchmarks noisyOr (F) (test)
0.002Inference Time (s)
6
PSI murderMystery (F) (test)
0Inference Time (s)
6
PSI grass (F) (test)
0.001Inference Time
6
PSI evidence2 (F) (test)
0Inference Time (s)
6
PSI evidence1 (F) benchmarks (test)
0Inference Time (s)
6
PSI digitRecognition (F) (test)
0.021Inference Time (s)
6
PSI alarm (F) (test)
0.001Inference Time (s)
6
PSI
0.023CIDEr
4
PSI clinicalTrial2 (C) (test)
0.002Inference Time (s)
3
PSI clinicalTrial (C) (test)
150Inference Time (s)
2
PSI 2.0 (test)
73Accuracy
1