Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

ViLP

Benchmarks

Task NameDataset NameSOTA ResultTrend
Self-evaluationViLP
AUROC77
36
Multimodal UnderstandingViLP 80% (test)
Accuracy58.06
10
Prior-Sensitive EvaluationViLP-F with-fact
Score70.17
8
Language Shortcut RobustnessViLP
Accuracy59.3
6
Uncertainty EstimationVILP
Verb-U0.521
2
Showing 5 of 5 rows