D3Former

Reported on 3 benchmarks across 1 task · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology3 results

Incremental LearningonCIFAR-100 - 50 classes + 25 steps of 2 classes
Average Incremental Accuracy· 2022-07-25
68.68
SOTA
D3Former: Debiased Dual Distilled Transformer for Incremental Learning arXiv:2208.00777
Incremental LearningonCIFAR-100 - 50 classes + 10 steps of 5 classes
Average Incremental Accuracy· 2022-07-25
70.94
best: 73.72 (TCIL)
D3Former: Debiased Dual Distilled Transformer for Incremental Learning arXiv:2208.00777
Incremental LearningonCIFAR-100 - 50 classes + 5 steps of 10 classes
Average Incremental Accuracy· 2022-07-25
72.23
best: 74.88 (TCIL)
D3Former: Debiased Dual Distilled Transformer for Incremental Learning arXiv:2208.00777