Audio-to-Text Generation on One-to-one evaluation benchmarks Audio-to-Text
55.11CIDErFlowBind
Evaluation results
| Method | Links | |
|---|---|---|
| FlowBindCategory=Generalists2025.12 | 55.11 | |
| OmniFlowCategory=Generalists2025.12 | 31.79 | |
| UnifiedIO2-LCategory=Generalists2025.12 | 12.15 | |
| CoDiCategory=Generalists2025.12 | 6.62 | |
| Qwen2-AudioCategory=Specialists2025.12 | 4.64 |