ASAP
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
ASAP (test)
94.59F1 Score
95
ASAP 1.0 (test)
0.836Prompt 1 QWK
51
ASAP n=102
64.3OMR-NED
16
ASAP 2.0
0.728QWK
16
ASAP
0.379QWK
15
ASAP Kaggle 2.0 (test)
0.84QWK
13
ASAP and ASAP++ (five-fold cross-val)
77.8Overall Score
11
ASAP and ASAP++ (five-fold cross-validation)
0.73Score P1
11
ASAP++ five-fold averaged results
0.712Overall Score
10
ASAP-SAS
0.661QWK (Prompt 3)
10
ASAP++ full-data setting
0.734Score P1
10
ASAP++ (full-data)
0.781Overall Score
10
ASAP (test)
0.764QWK
9
ASAP In-domain (5-fold cross-validation)
0.785Overall QWK
8
ASAP 2.0
49.51QWK
7
ASAP ACPAS definitions (test)
12.3Epsilon Onset Error
7
ASAP (test)
0.043Velocity JS Div
7
ASAP 7nm (test)
0.94Rise Delay MAPE
6
ASAP Prompt 8 (test)
0.694Ideas
6
ASAP Prompt 7 (test)
69.5Ideas Score
6
ASAP++
0.726QWK
5
ASAP
0.743QWK
5
ASAP (test)
65.1Content Score
4
ASAP Long Essays (Prompts 1, 2, 8)
83.6Score (P1)
4
(n)ASAP Dataset piano performances
6Mean Error (ms)
3