ResearchBenchmarksCross-modal classification (Audio to Visual) on VGGSound-AVEL UCF to VGG 40KFollow66.5PrecisionCoDAAR10.13224.76639.454.034May 12, 2026Evaluation ResultsMethodMethodLinksPrecisionCoDAARPretraining Scale=40KPretraining Scale=40K2026.0566.5MICUPretraining Scale=40KPretraining Scale=40K2026.0564.5DCIDPretraining Scale=40KPretraining Scale=40K2026.0560.6BaselinePretraining Scale=40KPretraining Scale=40K2026.0512.3