Class-Incremental Learning on AZ-Class IL
86.1Average AccuracyJoint
Evaluation Results
| Method | Links | |
|---|---|---|
| JointDescription=offline training on cumulative data; upper bound2026.05 | 86.1 | |
| FreeMOCALambda setting=Adap. Layerwise λ, Interpolation type=Linear2026.05 | 66.1 | |
| FreeMOCALambda setting=Adap. Layerwise λ, Interpolation type=Polynomial2026.05 | 64.2 | |
| FreeMOCALambda setting=Adap. Layerwise λ, Interpolation type=Spline2026.05 | 63.9 | |
| FreeMOCALambda setting=Fixed λ = 0.6, Interpolation type=Linear2026.05 | 63.7 | |
| FreeMOCALambda setting=Fixed λ = 0.6, Interpolation type=Spline2026.05 | 63.7 | |
| FreeMOCALambda setting=Fixed λ = 0.6, Interpolation type=Polynomial2026.05 | 63.7 | |
| CLeWI2026.05 | 62.2 | |
| MalCL2026.05 | 59 | |
| ER2026.05 | 56.9 | |
| WSC2026.05 | 55.7 | |
| TAMiL2026.05 | 54.8 | |
| iCaRL2026.05 | 54.7 | |
| NoneDescription=fine-tuning only; lower bound2026.05 | 26.5 | |
| GR2026.05 | 22.9 | |
| BI-R2026.05 | 22.5 | |
| EWC2026.05 | 17.1 | |
| LwF2026.05 | 10.6 | |
| SI2026.05 | 9.3 |