Video-to-Audio Retrieval on ACM
36.3Performance ScoreImageBind
Evaluation Results
| Method | Links | |
|---|---|---|
| ImageBindFusion=features fused as per UniIR2026.03 | 36.3 | |
| OmniRet2026.03 | 34.4 | |
| QwenOmni + GemmaDescription=multi-stage text-to-text retrieval pipeline2026.03 | 6.3 |