Text-to-Audio Retrieval on AudioCaps, Clotho, and MECAT Mean
0.221Recall@1M2D-CLAP
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| M2D-CLAPModel Category=CLAP Models2026.04 | 0.221 | 0.4786 | 0.5915 | |
| OEA-Nemo3BModel Category=OEA Models2026.04 | 0.2196 | 0.4591 | 0.5798 | |
| OEA-Qwen7BModel Category=OEA Models2026.04 | 0.2185 | 0.4677 | 0.5856 | |
| OEA-Qwen3B (+Cl)Model Category=OEA Models, Training Variant=+Cl2026.04 | 0.2154 | 0.4543 | 0.5667 | |
| OEA-Nemo3B (+Cl)Model Category=OEA Models, Training Variant=+Cl2026.04 | 0.2143 | 0.4626 | 0.5838 | |
| OEA-Qwen7B (+Cl)Model Category=OEA Models, Training Variant=+Cl2026.04 | 0.2127 | 0.464 | 0.5889 | |
| MGA-CLAPModel Category=CLAP Models2026.04 | 0.2106 | 0.458 | 0.5772 | |
| OEA-Qwen3BModel Category=OEA Models2026.04 | 0.2088 | 0.4372 | 0.5517 | |
| Robust-CLAPModel Category=CLAP Models2026.04 | 0.1594 | 0.3969 | 0.5232 | |
| LAION-CLAPModel Category=CLAP Models2026.04 | 0.141 | 0.3635 | 0.4894 | |
| Nemotron-3BModel Category=Vanilla LALMs2026.04 | 0.05 | 0.1464 | 0.2174 | |
| Qwen2.5-Omni-3BModel Category=Vanilla LALMs2026.04 | 0.0022 | 0.0082 | 0.0152 | |
| Qwen2.5-Omni-7BModel Category=Vanilla LALMs2026.04 | 0.0022 | 0.009 | 0.0201 |