Audio-Visual Captioning on DVD-Bench Zh
78AccuracyD-ORCA
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| D-ORCANumber of Parameters=8B2026.02 | 78 | 17.5 | 37.8 | |
| Ours-SFT ModelNumber of Parameters=8B2026.02 | 69.6 | 30.3 | 24.7 | |
| AVoCaDONumber of Parameters=7B2026.02 | 69.3 | — | — | |
| Qwen3-Omni-InstructNumber of Parameters=30B-A3B2026.02 | 63.5 | 60.6 | — | |
| ARC-Qwen-Video-NarratorNumber of Parameters=7B2026.02 | 63.2 | 53.6 | 10.1 | |
| video-SALMONN 2+Number of Parameters=7B2026.02 | 59.9 | — | — | |
| Qwen2.5-OmniNumber of Parameters=7B2026.02 | 55.7 | 69.4 | — |