Vision Tasks
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Image Classification | 14 Vision Tasks | Average Accuracy94.3 | 160 | |
| Visual Classification | 8 Vision Tasks (SUN397, Stanford Cars, RESISC45, EuroSAT, SVHN, GTSRB, MNIST, DTD) | Average Accuracy95.5 | 90 | |
| Image Classification | 8 Vision Tasks (test) | Avg Accuracy94.6 | 82 | |
| Model Merging | 8 Vision tasks (test) | Accuracy95.8 | 77 | |
| Image Classification | 8 vision tasks Average | Average Accuracy95.81 | 53 | |
| Multi-task Image Classification | 20 Vision Tasks | Accuracy94.8 | 48 | |
| Multi-task Image Classification | 14 vision tasks | Accuracy94.3 | 48 | |
| Multi-task Image Classification | 8 vision tasks | Accuracy95.9 | 48 | |
| Multi-task model merging | 14 vision tasks | Average Absolute Accuracy94.3 | 42 | |
| Multi-task model merging | 20 vision tasks | Average Absolute Accuracy94.7 | 40 | |
| 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 | |
| Multitask Learning | 8 vision tasks | Accuracy92.4 | 22 | |
| Model Merging | 8 Vision Tasks (average) | Average Accuracy86.3 | 18 | |
| Image Classification | 21 Vision Tasks (test) | Accuracy81.13 | 15 | |
| Multi-task Image Classification | 30 Vision Tasks | Accuracy93.1 | 12 | |
| Data Valuation | Vision Tasks | Kendall's Tau0.841 | 6 | |
| Multitask Learning | 14 vision tasks | Accuracy79.5 | 5 | |
| Image Classification | 9 Vision Tasks ImageNet-1K added (test) | SUN397 Accuracy77 | 5 | |
| Multitask Learning | 20 vision tasks | Accuracy72.9 | 4 |
Showing 20 of 20 rows