Algorithmic Recourse on GERMAN
77.1AccuracyStandard train → MixedPrec
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Standard train → MixedPrecTraining Objective=Standard, Quantization Strategy=MixedPrec2026.05 | 77.1 | 0.11 | 0.151 | |
| R-Margin train → MixedPrecTraining Objective=R-Margin, Quantization Strategy=MixedPrec2026.05 | 77.1 | 0.094 | 0.128 | |
| R-Consistency train → MixedPrecTraining Objective=R-Consistency, Quantization Strategy=MixedPrec2026.05 | 77.1 | 0.09 | 0.124 | |
| CFQ (ours, MixedPrec + CF objective)Training Objective=CF objective, Quantization Strategy=MixedPrec2026.05 | 77.1 | 0.052 | 0.064 |