WAKD (DeiT-Ti)

Reported on 4 benchmarks across 2 tasks · 1 paper

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

Methodology2 results

Domain AdaptationonPACS
Average Accuracy· 2023-09-20
87.6
best: 99 (SIMPLE+)
Weight Averaging Improves Knowledge Distillation under Domain Shift arXiv:2309.11446
Domain AdaptationonOffice-Home
Average Accuracy· 2023-09-20
70.5
best: 90.6 (MoA (OpenCLIP, ViT-B/16))
Weight Averaging Improves Knowledge Distillation under Domain Shift arXiv:2309.11446

Computer Vision2 results

Domain GeneralizationonPACS
Average Accuracy· 2023-09-20
87.6
best: 99 (SIMPLE+)
Weight Averaging Improves Knowledge Distillation under Domain Shift arXiv:2309.11446
Domain GeneralizationonOffice-Home
Average Accuracy· 2023-09-20
70.5
best: 90.6 (MoA (OpenCLIP, ViT-B/16))
Weight Averaging Improves Knowledge Distillation under Domain Shift arXiv:2309.11446