Audio-Visual QA on AVHBench
73.78AccuracyQwen2.5-Omni7B + OutRo
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2.5-Omni7B + OutRoBase Model=Qwen2.5-Omni7B, OutRo=true2026.03 | 73.78 | |
| Qwen2.5-Omni3B + OutRoBase Model=Qwen2.5-Omni3B, OutRo=true2026.03 | 71.67 | |
| Qwen2.5-Omni7BBase Model=Qwen2.5-Omni7B, OutRo=false2026.03 | 71.6 | |
| Qwen2.5-Omni3BBase Model=Qwen2.5-Omni3B, OutRo=false2026.03 | 70.91 | |
| video-SALMONN2+3B + OutRoBase Model=video-SALMONN2+3B, OutRo=true2026.03 | 62.33 | |
| video-SALMONN2+3BBase Model=video-SALMONN2+3B, OutRo=false2026.03 | 62.07 |