Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

POPE-GQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Object presence hallucination evaluationPOPE GQA Popular 2019
Accuracy78
8
Object Hallucination DiscriminationPOPE-GQA Adversarial
Accuracy75.17
6
Showing 2 of 2 rows