Audio-to-Text Retrieval on AVE
33.1AccuracyBrokenBind
Evaluation Results
| Method | Links | |
|---|---|---|
| BrokenBindBackbone=VCLIP+CLAP, Evaluation Protocol=BrokenBind, Source Dataset=MSRVTT2026.02 | 33.1 | |
| VCLIP+CLAP+FTBackbone=VCLIP+CLAP, Evaluation Protocol=Fine-tuning, Source Dataset=MSRVTT2026.02 | 17.2 | |
| VCLIP+CLAPBackbone=VCLIP+CLAP, Evaluation Protocol=Zero-shot, Source Dataset=MSRVTT2026.02 | 13.2 |