Image Classification on DTD (Zero-Shot Accuracy, CovGap, MCCC)
0.11CovGapTOPK
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| TOPKBackbone=RN101, Alpha=0.12026.06 | 0.11 | — | 0.07 | |
| L-TOPK (naive)Backbone=RN101, Alpha=0.12026.06 | 0.11 | — | 0.06 | |
| L-TOPKBackbone=RN101, Alpha=0.12026.06 | 0.11 | — | 0.19 | |
| LACBackbone=RN101, Alpha=0.12026.06 | 0.11 | — | 0.15 | |
| L-LAC (naive)Backbone=RN101, Alpha=0.12026.06 | 0.11 | — | 0.14 | |
| L-LACBackbone=RN101, Alpha=0.12026.06 | 0.11 | — | 0.22 | |
| APSBackbone=RN101, Alpha=0.12026.06 | 0.11 | — | 0.19 | |
| L-APS (naive)Backbone=RN101, Alpha=0.12026.06 | 0.11 | — | 0.18 | |
| L-APSBackbone=RN101, Alpha=0.12026.06 | 0.11 | — | 0.28 | |
| RAPSBackbone=RN101, Alpha=0.12026.06 | 0.12 | — | 0.08 | |
| L-RAPS (naive)Backbone=RN101, Alpha=0.12026.06 | 0.12 | — | 0.08 | |
| L-RAPSBackbone=RN101, Alpha=0.12026.06 | 0.12 | — | 0.11 | |
| Zero-Shot BaselineBackbone=RN1012026.06 | — | 37.1 | — |