| Paper | Code | Novel accuracy (50% Labeled) | Seen accuracy (50% Labeled) | All accuracy (50% Labeled) | All accuracy (10% Labeled) | Novel accuracy (10% Labeled) | Seen accuracy (10% Labeled) | ModelName | ReleaseDate |
|---|---|---|---|---|---|---|---|---|---|
| Parametric Classification for Generalized Category Discovery: A Baseline Study | ✓ Link | 79.1 | 92.4 | 83.6 | SimGCD (ViT-B-16) | 2022-11-21 | |||
| Open-World Semi-Supervised Learning | ✓ Link | 72.1 | 89.1 | 77.8 | 69.7 | 60.5 | 83.9 | ORCA (ResNet-50) | 2021-02-06 |
| OpenLDN: Learning to Discover Novel Classes for Open-World Semi-Supervised Learning | ✓ Link | 68.6 | 89.6 | 79.1 | OpenLDN (ResNet-50) | 2022-07-05 | |||
| Generalized Category Discovery | ✓ Link | 66.3 | 89.8 | 74.1 | GCD (ViT-B-16) | 2022-01-07 | |||
| Towards Realistic Semi-Supervised Learning | ✓ Link | 75.4 | 67.8 | 82.6 | TRSSL (ResNet-50) | 2022-07-05 |