Text-to-Audio Retrieval on FMACaps Eval
13Recall@1FIGMA
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| FIGMA2026.06 | 13 | 28 | 37.6 | 48.6 | |
| FIGMA2026.06 | 13 | 28 | 37.6 | 48.6 | |
| CLAMP 32026.06 | 7.5 | 20.7 | 30.8 | 43.1 | |
| LAION-CLAPVariant=Continued Training on FGMCaps2026.06 | 6.1 | 18.3 | 26.5 | 36.8 | |
| LAION-CLAPTraining Mode=Continued Training on FGMCaps2026.06 | 6.1 | 18.3 | 26.5 | 36.8 | |
| MuQ-MuLaN2026.06 | 4.1 | 12.4 | 17.8 | 27.5 | |
| LAION-CLAPVariant=Music, Speech and General Audio2026.06 | 3 | 7.5 | 12.9 | 20.8 | |
| LAION-CLAPTraining Data=Music, Speech and General Audio2026.06 | 3 | 7.5 | 12.9 | 20.8 | |
| LAION-CLAPVariant=Music and Speech2026.06 | 2.8 | 9.5 | 14.1 | 20.2 | |
| LAION-CLAPTraining Data=Music and Speech2026.06 | 2.8 | 9.5 | 14.1 | 20.2 | |
| LAION-CLAPVariant=Music2026.06 | 2.6 | 9.4 | 14.8 | 21.6 | |
| LAION-CLAPTraining Data=Music2026.06 | 2.6 | 9.4 | 14.8 | 21.6 | |
| LAION-CLAPVariant=General Audio2026.06 | 1.9 | 7.4 | 11.3 | 18.6 | |
| LAION-CLAPTraining Data=General Audio2026.06 | 1.9 | 7.4 | 11.3 | 18.6 | |
| M2D-CLAP2026.06 | 1.9 | 6.7 | 11.4 | 17.8 | |
| MS-CLAPYear=20232026.06 | 1.5 | 4.9 | 9.3 | 14.8 | |
| LAION-CLAPVariant=General audio with variable-length2026.06 | 0.7 | 5.1 | 9 | 17.4 | |
| LAION-CLAPTraining Data=General audio with variable-length2026.06 | 0.7 | 5.1 | 9 | 17.4 | |
| MS-CLAPYear=20222026.06 | 0.6 | 3.3 | 4.6 | 8.7 |