Image Classification on Generalization Suite Unseen EuroSAT MNIST ViT-B/32 (Split 1)
86.3Cars AccuracyTARA-Variant B
Evaluation Results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| TARA-Variant BBackbone=ViT-B/32, Normalization=to finetuned (%)2026.03 | 86.3 | 73.8 | 47.6 | 73.8 | 99.1 | 71.6 | 75.4 | 53.3 | 61.6 | 57.5 | 70.9 | |
| TARA-Variant ABackbone=ViT-B/32, Normalization=to finetuned (%)2026.03 | 82.8 | 73.9 | 43.8 | 70.5 | 98.2 | 73.4 | 73.8 | 48.4 | 61.1 | 54.8 | 69.1 | |
| KnOTS-TIESBackbone=ViT-B/32, Normalization=to finetuned (%)2026.03 | 82.7 | 74.4 | 48.5 | 72.5 | 95.3 | 50.2 | 70.6 | 33.2 | 50.5 | 41.8 | 63.4 | |
| LoRA-LEGOBackbone=ViT-B/32, Normalization=to finetuned (%)2026.03 | 82.2 | 73.5 | 46.9 | 71.2 | 96.7 | 40.8 | 68.5 | 35.5 | 44.7 | 40.1 | 61.4 | |
| TABackbone=ViT-B/32, Normalization=to finetuned (%)2026.03 | 82 | 74.7 | 42.2 | 71.1 | 96.8 | 42.2 | 68.2 | 37.5 | 48.2 | 42.9 | 61.8 | |
| KnOTS-DARE-TIESBackbone=ViT-B/32, Normalization=to finetuned (%)2026.03 | 81.7 | 72.5 | 49.4 | 71.9 | 94.6 | 52.5 | 70.4 | 31.6 | 51.2 | 41.4 | 63.2 | |
| TIESBackbone=ViT-B/32, Normalization=to finetuned (%)2026.03 | 81.5 | 73.2 | 40.3 | 64.4 | 95.4 | 45.7 | 66.7 | 37.2 | 61.8 | 49.5 | 62.4 | |
| AdaMergingBackbone=ViT-B/32, Normalization=to finetuned (%)2026.03 | 79.7 | 73.4 | 37.7 | 69.8 | 97.9 | 67.4 | 71 | 48.7 | 58.7 | 53.7 | 66.7 |