Vision Tasks
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Image Classification | 14 Vision Tasks | Average Accuracy94.3 | 84 | |
| Image Classification | 8 Vision Tasks (test) | Avg Accuracy94.6 | 82 | |
| Visual Classification | 8 Vision Tasks (SUN397, Stanford Cars, RESISC45, EuroSAT, SVHN, GTSRB, MNIST, DTD) | Average Accuracy95.5 | 60 | |
| Model Merging | 8 Vision tasks (test) | Accuracy89 | 33 | |
| Multi-task Image Classification | 8 Vision Tasks Low Confidence setting (test) | Absolute Accuracy94 | 32 | |
| Multi-task Image Classification | 8 Vision Tasks Norm Mismatch setting (test) | Absolute Accuracy93.3 | 32 | |
| Image Classification | 9 Vision Tasks ImageNet-1K added (test) | SUN397 Accuracy77 | 5 |
Showing 7 of 7 rows