Continual Anomaly Detection on Image field tabular datasets
76.91Balanced AccuracyTaskFusion (AGF+OE+WT+CT+DD)
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| TaskFusion (AGF+OE+WT+CT+DD)Replay strategy=Dataset Distillation, OE=true, WT=true, CT=true2026.06 | 76.91 | 52.83 | 85.2 | |
| AGF+OE+WT+DDReplay strategy=Dataset Distillation, OE=true, WT=true2026.06 | 75.87 | 52.44 | 84.65 | |
| AGF+OE+CT+DDReplay strategy=Dataset Distillation, OE=true, CT=true2026.06 | 75.29 | 52.12 | 84.13 | |
| AGF+OE+DDReplay strategy=Dataset Distillation, Outlier Exposure (OE)=true2026.06 | 74.72 | 51.14 | 84.72 | |
| MultitaskContinual Learning protocol=Joint2026.06 | 72.8 | 55.49 | 85.06 | |
| AGF+DDReplay strategy=Dataset Distillation2026.06 | 72.06 | 50.32 | 84.6 | |
| Tasks w/o CLContinual Learning protocol=w/o CL2026.06 | 71.55 | 54.74 | 84 | |
| AGF+RSReplay strategy=Random Sampling2026.06 | 69.54 | 46.48 | 80.86 | |
| EDSR2026.06 | 62.75 | 40.26 | 74.91 | |
| CaSSLe2026.06 | 54.27 | 27.01 | 65.94 | |
| FinetuneContinual Learning protocol=CL2026.06 | 53.02 | 32.96 | 59.43 |