| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Image Classification | STL10 | Accuracy99.6 | 60 | |
| Image Classification | STL10 (test) | Error Rate (%)4.82 | 53 | |
| Image Classification | STL10-LT (gamma_l = 10) (test) | Accuracy84.2 | 42 | |
| Image Classification | STL10-LT (test) | Accuracy84.8 | 36 | |
| Image Classification | STL10 (val) | Accuracy98 | 33 | |
| Image Classification | STL10 gamma_l = 20 long-tail (test) | Accuracy82.3 | 26 | |
| Semi-supervised Long-tailed Image Classification | STL10 LT | Accuracy79 | 24 | |
| Image Classification | STL10 Dp: ImageNet (test) | BA82.94 | 22 | |
| Debugging Effectiveness Estimation | STL10 | ATPF88.67 | 21 | |
| Clustering | STL10 small-scale (val) | NMI99.2 | 17 | |
| Image Clustering | STL10 (labelled) | Accuracy59.8 | 17 | |
| Image Classification | STL10 (Nl=150, M=100k, γl=10) long-tailed (test) | Accuracy71.3 | 16 | |
| Image Classification | STL10 43 (test) | Accuracy77.45 | 15 | |
| Image Generation | STL10 | Inception Score10.41 | 15 | |
| Generative Modeling | STL10 96x96 (test) | FID77.24 | 12 | |
| Model Stealing Attack | STL10 | Agreement94.2 | 11 | |
| Adversarial Robustness | STL10 (test) | Baseline Acc82.07 | 10 | |
| Zero-shot Classification | STL10 | Top-1 Clean Acc97.6 | 10 | |
| Attacking Non-Transferable Learning | STL10 to CIFAR10 (test) | Accuracy (Authorized Domain)88.3 | 10 | |
| Image Classification | STL10 C8-augmented (test) | C8-Avg Accuracy96.37 | 10 | |
| Image Classification | STL10 original (test) | Accuracy98.31 | 10 | |
| Unsupervised image clustering | STL10 (unlabeled) | Accuracy59.6 | 9 | |
| Image Classification | STL10 downstream | Classification Accuracy95.68 | 8 | |
| Image Classification | STL10-scale non-ideal downsampling | Accuracy72.62 | 8 | |
| Image Classification | STL10 scale (test) | Accuracy73.32 | 8 |