ResearchBenchmarksScene & Action Recognition on UCF101 T=20Follow98.71Accuracy (AT)VILA97.690897.955498.2298.4846Feb 14, 2026Evaluation ResultsMethodMethodLinksAccuracy (AT)Accuracy (A_bar)VILABackbone=ViT-B/16Backbone=ViT-B/162026.0298.7199.35RanPACBackbone=ViT-B/16Backbone=ViT-B/162026.0297.7398.83