Audio-to-Text Retrieval on FMACaps Eval
13.2R@1FIGMA
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| FIGMA2026.06 | 13.2 | 33.3 | 42.9 | 53.4 | |
| FIGMA2026.06 | 13.2 | 33.3 | 42.9 | 53.4 | |
| LAION-CLAPVariant=Continued Training on FGMCaps2026.06 | 6 | 20 | 30.1 | 40.9 | |
| LAION-CLAPTraining Mode=Continued Training on FGMCaps2026.06 | 6 | 20 | 30.1 | 40.9 | |
| LAION-CLAPVariant=General Audio2026.06 | 4 | 12.1 | 19.1 | 27.6 | |
| LAION-CLAPTraining Data=General Audio2026.06 | 4 | 12.1 | 19.1 | 27.6 | |
| MuQ-MuLaN2026.06 | 3.9 | 11.7 | 19.1 | 28.1 | |
| M2D-CLAP2026.06 | 3.3 | 11.9 | 19.7 | 30.8 | |
| LAION-CLAPVariant=Music2026.06 | 3 | 11.6 | 18.5 | 28.1 | |
| LAION-CLAPTraining Data=Music2026.06 | 3 | 11.6 | 18.5 | 28.1 | |
| MS-CLAPYear=20232026.06 | 2.9 | 8.3 | 15.5 | 22.9 | |
| LAION-CLAPVariant=Music, Speech and General Audio2026.06 | 2.8 | 10.1 | 17.9 | 28 | |
| LAION-CLAPTraining Data=Music, Speech and General Audio2026.06 | 2.8 | 10.1 | 17.9 | 28 | |
| LAION-CLAPVariant=Music and Speech2026.06 | 2.4 | 9.8 | 17.4 | 25.5 | |
| LAION-CLAPTraining Data=Music and Speech2026.06 | 2.4 | 9.8 | 17.4 | 25.5 | |
| LAION-CLAPVariant=General audio with variable-length2026.06 | 2.2 | 7.3 | 11.5 | 21.1 | |
| LAION-CLAPTraining Data=General audio with variable-length2026.06 | 2.2 | 7.3 | 11.5 | 21.1 | |
| CLAMP 32026.06 | 1.1 | 4.1 | 6.4 | 11.7 | |
| MS-CLAPYear=20222026.06 | 0.8 | 3.4 | 6.8 | 12 |