Audio Classification on ESC50 Actions
100Accuracy (Base)SubT
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SubT2026.06 | 100 | 72.67 | 83.89 | |
| SubT†2026.06 | 100 | 76.5 | 86.44 | |
| COOPSetting=Few-Shot2026.06 | 100 | 72.5 | — | |
| COCOOPSetting=Few-Shot2026.06 | 100 | 62.5 | — | |
| CoCoOp2026.06 | 99.17 | 81.33 | 89.26 | |
| CoOp2026.06 | 99 | 83.67 | 90.43 | |
| DePT2026.06 | 98.67 | 88.33 | 93.15 | |
| SEPT2026.06 | 98.5 | 90.67 | 94.38 | |
| COCOOP + ZEBRASetting=Few-Shot, Framework=ZEBRA2026.06 | 97.5 | 77.5 | — | |
| CLIP-Adapter2026.06 | 96.33 | 70.17 | 80.37 | |
| COOP + ZEBRASetting=Few-Shot, Framework=ZEBRA2026.06 | 95 | 77.5 | — | |
| CoCoOpShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 89.17 | 41.5 | 56.15 | |
| CoCoOp + SEPTShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 87.5 | 51.5 | 64.58 | |
| CoOpShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 85.33 | 49.33 | 61.97 | |
| CoOplearning_mode=16-shot learning2026.01 | 85.33 | 49.33 | 61.97 | |
| Zero-shot2026.06 | 83.5 | 77.5 | 80.3 | |
| CoOp + SEPTShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 82.5 | 52.67 | 63.9 | |
| DePTShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 81.33 | 44.67 | 57.12 | |
| DePT + SEPTShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 79.67 | 52.67 | 63.02 | |
| KgCoOp2026.06 | 77.5 | 37.5 | 49.71 | |
| KgCoOp + SEPTShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 67.83 | 51.5 | 57.97 | |
| ZERO SHOTSetting=Zero-Shot2026.06 | 67.5 | 75 | — | |
| KgCoOp†learning_mode=16-shot learning2026.01 | 66 | 50.17 | 56.68 | |
| KgCoOp† + SEPTlearning_mode=16-shot learning2026.01 | 65.33 | 53.17 | 58.37 | |
| KgCoOpShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 36.5 | 30.67 | 32.88 | |
| KgCoOplearning_mode=16-shot learning2026.01 | 36.5 | 30.67 | 32.88 | |
| PengiShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 25.5 | 28 | 26.44 |