Audio-to-text Retrieval on AudioCaps (test)
65.6R@1AuroLA
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| AuroLARe-ranking=true2026.02 | 65.6 | 87.5 | 93.3 | — | — | |
| AuroLARe-ranking=true2026.02 | 65.6 | — | — | — | — | |
| AuroLA2026.02 | 64 | 86.7 | 92.9 | — | — | |
| AuroLA2026.02 | 64 | — | — | — | — | |
| SLAPEvaluation Protocol=Supervised2026.01 | 63.4 | 87.8 | 94.1 | — | — | |
| PE-AV2026.02 | 63.3 | — | — | — | — | |
| PE-AV2026.02 | 63.3 | — | — | — | — | |
| FineLAP2026.04 | 62.5 | 85.9 | 95 | — | — | |
| CED-LEEvaluation Protocol=Supervised2026.01 | 60.7 | 86.9 | 94.8 | — | — | |
| M2D2Evaluation Protocol=Supervised2026.01 | 59.2 | 84.7 | 92.7 | — | — | |
| M2D-CLAP2026.04 | 59.2 | 84.7 | 92.7 | — | — | |
| AuroLAPre-training (PT)=true, Re-ranking=true2026.02 | 58.7 | 84 | 90.3 | — | — | |
| AuroLAPre-training (PT)=true, Re-ranking=true2026.02 | 58.7 | — | — | — | — | |
| ML-CLAP2026.02 | 55.7 | 81.9 | 90.8 | — | — | |
| ML-CLAP2026.02 | 55.7 | — | — | — | — | |
| ALM2VecStage=FT2026.06 | 55.5 | 80 | 88.2 | — | — | |
| CacophonyEvaluation Protocol=Supervised2026.01 | 55.3 | 83.6 | 92.4 | — | — | |
| Cacophony2026.02 | 55.3 | 83.6 | 92.4 | — | — | |
| Cacophony2026.02 | 55.3 | — | — | — | — | |
| Cacophony2026.04 | 55.3 | 83.6 | 92.4 | — | — | |
| AuroLAPre-training (PT)=true2026.02 | 54.8 | 81.7 | 89.2 | — | — | |
| AuroLAPre-training (PT)=true2026.02 | 54.8 | — | — | — | — | |
| WavCapsEvaluation Protocol=Supervised2026.01 | 54.6 | 85.2 | 92.4 | — | — | |
| WavCaps-CLAPStage=FT2026.06 | 54.6 | 85.2 | 92.4 | — | — | |
| MGA-CLAP2026.04 | 53.7 | 84.3 | — | — | — | |
| FLAPmode=fusion2026.02 | 53 | 84.1 | 92.6 | — | — | |
| FLAPFusion=true2026.02 | 53 | — | — | — | — | |
| WavCaps2026.02 | 51.7 | 82.3 | 90.6 | — | — | |
| WavCaps2026.02 | 51.7 | — | — | — | — | |
| HTSAT-BERT2026.04 | 51.7 | 82.3 | 90.6 | — | — | |
| WavCaps-CLAPStage=PT2026.06 | 51.7 | 82.3 | 90.6 | — | — | |
| OnePeace2026.02 | 51 | 81.9 | 92 | — | — | |
| OnePeace2026.02 | 51 | — | — | — | — | |
| M-LTM2026.02 | 49.9 | 80.8 | 90.5 | — | — | |
| M-LTM2026.02 | 49.9 | — | — | — | — | |
| PE-AVPre-training (PT)=true, Reproduced=true2026.02 | 48.5 | 78.5 | 90.3 | — | — | |
| PE-AVPre-training (PT)=true2026.02 | 48.5 | — | — | — | — | |
| ReCLAP2026.02 | 48 | 80.4 | 90.8 | — | — | |
| ReCLAP2026.02 | 48 | — | — | — | — | |
| CompA-CLAP2026.02 | 47.8 | 83.5 | 90.2 | — | — | |
| CompA-CLAP2026.02 | 47.8 | — | — | — | — | |
| LAION-CLAP2026.06 | 46.8 | 82.9 | 90.7 | — | — | |
| LAION-CLAPEvaluation Protocol=Supervised2026.01 | 45 | 76.7 | 88 | — | — | |
| CLAPmode=fusion2026.02 | 45 | 76.7 | 88 | — | — | |
| CLAPFusion=true2026.02 | 45 | — | — | — | — | |
| SLAPEvaluation Protocol=Zero-Shot2026.01 | 44.8 | 73.9 | 84.4 | — | — | |
| WAVE 7B2025.09 | 44.2 | — | — | — | — | |
| LAION-CLAP2026.04 | 44.2 | 80.8 | 90.3 | — | — | |
| CLAP2026.02 | 43.9 | 77.7 | 87.6 | — | — | |
| CLAP2026.02 | 43.9 | — | — | — | — | |
| ALM2VecStage=PT2026.06 | 43.8 | 74.3 | 86.5 | — | — | |
| FLAM2026.04 | 43.3 | 75 | — | — | — | |
| DiffATR2026.02 | 42.6 | 74.4 | 86.6 | — | — | |
| DiffATR2026.02 | 42.6 | — | — | — | — | |
| Mei et al. (2024)Reference Type=Reference Model2025.09 | 42.2 | — | — | — | — | |
| WavCapsEvaluation Protocol=Zero-Shot2026.01 | 40.2 | 69.4 | 80.3 | — | — | |
| WavCapsCategory=BM2025.11 | 38.7 | 73.32 | 86.05 | 66.02 | — | |
| M2D2Evaluation Protocol=Zero-Shot2026.01 | 36.1 | 65.3 | 77.4 | — | — | |
| VASTPre-training (PT)=true, Reproduced=true2026.02 | 34.8 | 65.4 | 76.5 | — | — | |
| VASTPre-training (PT)=true, Reproduced=true2026.02 | 34.8 | — | — | — | — | |
| CLAP (LAION)Category=BM2025.11 | 34.58 | 70.8 | 83.69 | 63.02 | — | |
| CACARA_ASC/AA/WC/AC/C/f 0.2Category=CACARA, f-value=0.22025.11 | 33.98 | 68.3 | 81.81 | 61.36 | — | |
| CACARA_ASC/AA/WC/AC/C/f 0.1Category=CACARA, f-value=0.12025.11 | 33.64 | 68.4 | 81.84 | 61.29 | — | |
| CACARA_ASC/AA/WC/AC/CCategory=CACARA2025.11 | 33.27 | 67.46 | 81.51 | 60.75 | — | |
| MS-CLAP2026.06 | 32 | 66 | 79.2 | — | — | |
| CACARA_AA/WC/AC/CCategory=CACARA2025.11 | 31.03 | 64.57 | 79.34 | 58.31 | — | |
| CACARA_WC/AC/C/f 0.2Category=CACARA, f-value=0.22025.11 | 31.03 | 65.32 | 78.8 | 58.38 | — | |
| CACARA_WC/AC/C/f 0.1Category=CACARA, f-value=0.12025.11 | 30.96 | 64.78 | 78.54 | 58.1 | — | |
| CACARA_WC/AC/CCategory=CACARA2025.11 | 30.08 | 64.95 | 78.71 | 57.91 | — | |
| MixProLAPZero-shot=true, Training Dataset=AudioCaps2026.06 | 26.85 | — | 68.37 | — | 20.24 | |
| CLAPZero-shot=true, Training Dataset=AudioCaps2026.06 | 24.23 | — | 63.71 | — | 18.89 | |
| OnePeacePre-training (PT)=true, Reproduced=true2026.02 | 24 | 58.9 | 74.1 | — | — | |
| OnePeacePre-training (PT)=true, Reproduced=true2026.02 | 24 | — | — | — | — | |
| JINA-Embed.-v52026.06 | 23.1 | 52.7 | 67.2 | — | — | |
| MixProLAPZero-shot=true, Training Dataset=ClothoV22026.06 | 19.8 | — | 60.86 | — | 14.99 | |
| VASTCategory=MM2025.11 | 19.75 | 46.3 | 57.98 | 41.34 | — | |
| CLAPZero-shot=true, Training Dataset=ClothoV22026.06 | 19.68 | — | 55.52 | — | 13.4 | |
| CLAP (Microsoft)Category=BM2025.11 | 15.75 | 44.7 | 61.62 | 40.69 | — | |
| LanguageBindCategory=MM2025.11 | 13.05 | 45.96 | 63.76 | 40.92 | — | |
| ImageBindCategory=MM2025.11 | 8.59 | 27.58 | 40.49 | 25.55 | — | |
| LAION-CLAPParams=355 M + 31 M2026.04 | — | 46.7 | 65 | — | — | |
| Woosh-CLAP-PrivateParams=355 M + 86 M2026.04 | — | 15.3 | 22.7 | — | — | |
| Woosh-CLAP-PublicParams=355 M + 86 M2026.04 | — | 46.2 | 59.5 | — | — |