Audio-to-Text Retrieval (via Vision pivot) on MSRVTT to AVE (test)
29.2mAPLB+BB
Evaluation results
| Method | Links | |
|---|---|---|
| LB+BBBackbone=LanguageBind, Variant=BrokenBind2026.02 | 29.2 | |
| IB+BBBackbone=ImageBind, Variant=BrokenBind2026.02 | 27.8 | |
| LB+FTBackbone=LanguageBind, Variant=Fine-tuned2026.02 | 13.6 | |
| LBBackbone=LanguageBind, Variant=Zero-shot2026.02 | 11.2 | |
| IB+FTBackbone=ImageBind, Variant=Fine-tuned2026.02 | 9.3 | |
| IBBackbone=ImageBind, Variant=Zero-shot2026.02 | 8.1 |