Model Selection on LOVM average over 23 datasets
0.504R5 ScoreSWAB
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SWABParadigm=Learning-based2024.03 | 0.504 | 0.318 | 0.822 | |
| ModelGPTParadigm=Learning-based2024.03 | 0.446 | 0.27 | 0.716 | |
| INBParadigm=Open-source dataset-based, Reference Dataset=ImageNet2024.03 | 0.443 | 0.267 | 0.71 | |
| Avg RankParadigm=Open-source dataset-based2024.03 | 0.443 | 0.246 | 0.689 | |
| NCEParadigm=Metric-based (Classic)2024.03 | 0.235 | -0.014 | 0.221 | |
| LogMEParadigm=Metric-based (Classic)2024.03 | 0.191 | -0.014 | 0.177 | |
| H-ScoreParadigm=Metric-based (Classic)2024.03 | 0.174 | 0 | 0.174 | |
| LEEPParadigm=Metric-based (Classic)2024.03 | 0.161 | 0.014 | 0.175 |