Audio Classification on Beijing-Opera
100Base AccuracyCoCoOp
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| CoCoOp2026.06 | 100 | 52.53 | 68.74 | |
| SubT2026.06 | 100 | 60.63 | 74.72 | |
| SubT†2026.06 | 100 | 52.26 | 68.58 | |
| CLIP-Adapter2026.06 | 99.68 | 51.2 | 67.65 | |
| CoOp + SEPTShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 97.88 | 71.87 | 82.45 | |
| CoCoOpShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 97.86 | 70.8 | 81.84 | |
| CoCoOp + SEPTShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 97.85 | 71.06 | 81.6 | |
| CoOp2026.06 | 97.78 | 51.47 | 67.31 | |
| KgCoOp†learning_mode=16-shot learning, prompt_templates=100 hand-crafted ensemble2026.01 | 97.53 | 72.94 | 82.55 | |
| CoOpShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 97.27 | 61.38 | 74.87 | |
| CoOplearning_mode=16-shot learning2026.01 | 97.27 | 61.38 | 74.87 | |
| DePTShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 97.26 | 62.25 | 75.32 | |
| KgCoOp† + SEPTlearning_mode=16-shot learning, prompt_templates=100 hand-crafted ensemble2026.01 | 97.23 | 71.22 | 81.41 | |
| DePT + SEPTShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 96.32 | 62.74 | 75 | |
| COCOOP + ZEBRASetting=Few-Shot, Framework=ZEBRA2026.06 | 96.2 | 78.26 | — | |
| COOPSetting=Few-Shot2026.06 | 96.1 | 60.87 | — | |
| COOP + ZEBRASetting=Few-Shot, Framework=ZEBRA2026.06 | 96.03 | 82.61 | — | |
| COCOOPSetting=Few-Shot2026.06 | 96 | 60.88 | — | |
| KgCoOp + SEPTShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 94.44 | 67.99 | 78.28 | |
| SEPT2026.06 | 89.87 | 46.94 | 60.71 | |
| DePT2026.06 | 86.87 | 50.48 | 61.79 | |
| Zero-shot2026.06 | 79.74 | 51.2 | 62.2 | |
| LTU-7Br=8, Conn.=Linear2025.09 | 69.9 | — | — | |
| GAMA-7Br=8, Conn.=QFormer2025.09 | 69.5 | — | — | |
| SAM-2.7Br=8, Conn.=(a), Experiment=E32025.09 | 69.1 | — | — | |
| KgCoOpShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 67.26 | 61.4 | 62.96 | |
| KgCoOplearning_mode=16-shot learning, prompt_templates=single2026.01 | 67.26 | 61.4 | 62.96 | |
| KgCoOp2026.06 | 65.52 | 50.94 | 56.2 | |
| SAM-2.7Br=256, Conn.=(a), Experiment=E62025.09 | 65.3 | — | — | |
| SAM-2.7Br=256, Conn.=(b), Ablation=4.2, Experiment=E142025.09 | 64.4 | — | — | |
| SAM-2.7Br=256, Conn.=(c), Ablation=4.2, Experiment=E152025.09 | 64.4 | — | — | |
| PengiShots=16, Evaluation Protocol=Base-to-new generalization2026.01 | 64.16 | 68.55 | 66.15 | |
| SAM-130Mr=256, Conn.=(c), Ablation=4.2, Experiment=E112025.09 | 62.7 | — | — | |
| SAM-130Mr=256, Conn.=(a), Experiment=E42025.09 | 62.3 | — | — | |
| ssLALM-2.8Br=8, Conn.=Linear2025.09 | 61 | — | — | |
| SAM-780Mr=256, Conn.=(b), Ablation=4.2, Experiment=E122025.09 | 61 | — | — | |
| SAM-2.7Br=256, Conn.=(a), Ablation=4.1, Experiment=E92025.09 | 58.9 | — | — | |
| SAM-780Mr=256, Conn.=(c), Ablation=4.2, Experiment=E132025.09 | 58.1 | — | — | |
| SAM-780Mr=8, Conn.=(a), Experiment=E22025.09 | 57.6 | — | — | |
| SAM-130Mr=256, Conn.=(a), Ablation=4.3, Experiment=E162025.09 | 57.6 | — | — | |
| SAM-2.7Br=256, Conn.=(a), Ablation=4.3, Experiment=E182025.09 | 57.2 | — | — | |
| SAM-130Mr=256, Conn.=(a), Ablation=4.1, Experiment=E72025.09 | 56.8 | — | — | |
| SAM-780Mr=256, Conn.=(a), Ablation=4.1, Experiment=E82025.09 | 52.5 | — | — | |
| ZERO SHOTSetting=Zero-Shot2026.06 | 52 | 43.48 | — | |
| SAM-130Mr=8, Conn.=(a), Experiment=E12025.09 | 49.2 | — | — | |
| SAM-780Mr=256, Conn.=(a), Experiment=E52025.09 | 49.2 | — | — | |
| SAM-780Mr=256, Conn.=(a), Ablation=4.3, Experiment=E172025.09 | 48.3 | — | — | |
| SAM-130Mr=256, Conn.=(b), Ablation=4.2, Experiment=E102025.09 | 46.6 | — | — |