Model Ratatouille

Reported on 6 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.

Methodology3 results

Domain AdaptationonPACS
Average Accuracy· 2022-12-20
90.5
best: 99 (SIMPLE+)
Model Ratatouille: Recycling Diverse Models for Out-of-Distribution Generalization arXiv:2212.10445
Domain AdaptationonOffice-Home
Average Accuracy· 2022-12-20
73.5
best: 90.6 (MoA (OpenCLIP, ViT-B/16))
Model Ratatouille: Recycling Diverse Models for Out-of-Distribution Generalization arXiv:2212.10445
Domain AdaptationonTerraIncognita
Average Accuracy· 2022-12-20
52
best: 69.6 (UniDG + CORAL + ConvNeXt-B)
Model Ratatouille: Recycling Diverse Models for Out-of-Distribution Generalization arXiv:2212.10445

Computer Vision3 results

Domain GeneralizationonPACS
Average Accuracy· 2022-12-20
90.5
best: 99 (SIMPLE+)
Model Ratatouille: Recycling Diverse Models for Out-of-Distribution Generalization arXiv:2212.10445
Domain GeneralizationonOffice-Home
Average Accuracy· 2022-12-20
73.5
best: 90.6 (MoA (OpenCLIP, ViT-B/16))
Model Ratatouille: Recycling Diverse Models for Out-of-Distribution Generalization arXiv:2212.10445
Domain GeneralizationonTerraIncognita
Average Accuracy· 2022-12-20
52
best: 69.6 (UniDG + CORAL + ConvNeXt-B)
Model Ratatouille: Recycling Diverse Models for Out-of-Distribution Generalization arXiv:2212.10445