ResearchBenchmarksAudio-Visual Captioning on VGGSound All (with ALOHa)Follow38.8ALOHaASD32.24833.94935.6537.351May 11, 2026Evaluation ResultsMethodMethodLinksALOHaCsCiF1 ScoreASDBase Model=Qwen2.5-Omn...Base Model=Qwen2.5-Omni(7B)2026.0538.829.621.755.8VanillaBase Model=Qwen2.5-Omn...Base Model=Qwen2.5-Omni(7B)2026.053530.720.658.6AVCDBase Model=Qwen2.5-Omn...Base Model=Qwen2.5-Omni(7B)2026.0532.833.524.455.7FMDBase Model=Qwen2.5-Omn...Base Model=Qwen2.5-Omni(7B)2026.0532.537.925.957.3