| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Image Classification | CINIC-10 (test) | Accuracy95.8 | 177 | |
| Image Classification | CINIC-10 | Accuracy65 | 59 | |
| Image Classification | CINIC-10 1.0 (10% label) | Accuracy81.29 | 42 | |
| Image Classification | CINIC-10 Dir(0.01), 50 clients, 20% participation | Accuracy56.96 | 36 | |
| Image Classification | CINIC-10 Dir(0.5) (test) | Accuracy62.56 | 28 | |
| Image Classification | CINIC-10 Dir(0.5) | Accuracy62.96 | 28 | |
| Image Classification | CINIC-10 under sudden drift and Dir(0.1) 1.0 (test) | Generalized Accuracy46.02 | 27 | |
| Image Classification | CINIC-10 non-iid | Accuracy37.59 | 26 | |
| Image Classification | CINIC-10 iid (test) | Test Accuracy40.8 | 26 | |
| Image Classification | CINIC-10 20 clients (test) | Accuracy51.12 | 14 | |
| Poisoning Defense in U-shape Split Learning | CINIC-10 | Accuracy67 | 10 | |
| Predicting Generalization | CINIC FCN PGDL tasks (train test) | CMI Score16.93 | 10 | |
| Online Label Shift | CINIC10 Bernoulli shift | Average Error25.63 | 7 | |
| Online Label Shift | CINIC10 Linear shift | Average Error26.21 | 7 | |
| Online Label Shift Adaptation | CINIC10 Square shift (test) | Average Error25.56 | 7 | |
| Federated Learning Efficiency | CINIC-10 (test) | Communication Rounds104 | 6 | |
| Poison Defense | CINIC-10 (test) | Avg Poison Success4.76 | 6 | |
| VFL Attack Performance | CINIC-10 | LISR17.82 | 4 | |
| Generalization Gap Prediction | CINIC-10 | CMI Score33.76 | 4 | |
| Accuracy Estimation | CINIC-10 (test) | AUPR86.45 | 4 | |
| Clustering | CINIC-10 (test) | Accuracy43.06 | 4 |