List your company

Audio-Visual Captioning on VGGSound Animal

51.52Cs ScoreVCD
24.01231.153538.29545.4365May 11, 2026

Evaluation results

MethodLinks
VCDModel=Qwen2.5-Omni(7B)2026.0551.5241.2852.4340.27
AVCDBase Model=Qwen2.5-Omni(7B)2026.0551.538.853.738.5
VanillaModel=Qwen2.5-Omni(3B)2026.0551.244152.88—
PAIModel=Qwen2.5-Omni(7B)2026.0551.2438.1155.1139.52
FMDBase Model=Qwen2.5-Omni(7B)2026.0550.940.252.937.3
VanillaModel=Qwen2.5-Omni(7B)2026.0548.2137.1355.2440.71
VanillaBase Model=Qwen2.5-Omni(7B)2026.0548.237.155.240.7
ASDModel=Qwen2.5-Omni(3B)2026.0543.2538.5250.44—
VCDModel=video-SALMONN-o1 (7B)2026.0539.3933.453.3736.4
VanillaModel=video-SALMONN-o1 (7B)2026.0537.7432.0953.6836.21
ASDModel=Qwen2.5-Omni(7B)2026.0536.9134.1552.4442.77
ASDBase Model=Qwen2.5-Omni(7B)2026.0536.934.152.442.7
PAIModel=video-SALMONN-o1 (7B)2026.0535.2631.1853.1636.99
ASDModel=video-SALMONN-o1 (7B)2026.0525.0725.7150.8943.29