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

CVQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Counterfactual ReasoningCVQA
Accuracy71.41
40
Visual Question AnsweringCVQA
Global Accuracy (CVQA)67.59
14
Visual Question AnsweringcVQA
Accuracy86.6
12
Causal Variable IdentificationCVQA Count
F1 (X)74.7
7
Causal Variable IdentificationCVQA Bool
F1 (X)79.8
7
Outcome ReasoningCVQA Count
F1 Mean (M')0.792
7
Outcome ReasoningCVQA-Bool
M' (F1 Score)81.2
7
Showing 7 of 7 rows