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

P2

Benchmarks

Task NameDataset NameSOTA ResultTrend
Face Forgery DetectionP2 Hybrid, FR, FS, EFS v1 (test)
Hybrid Score97
40
Prompt OptimizationP2-hard
DSGScore92
7
Evolutionary Multi-Task OptimizationP2
Normalized Fitness1.036
6
Bias MitigationP2 Race unconditional 1.0
FD0.317
5
Bias MitigationP2 Age unconditional 1.0
FD0.401
5
Bias MitigationP2 Gender 1.0 (unconditional)
FD0.002
5
Investment decision alignmentP2 v1 (test)
Overall MSE6.12
4
Showing 7 of 7 rows