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

GQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Visual Question AnsweringGQA
Accuracy81.9
1,249
Visual Question AnsweringGQA
Accuracy74.9
505
Visual Question AnsweringGQA
Mean Accuracy65.8
196
Visual Question AnsweringGQA
Score71.3
193
Visual Question AnsweringGQA (test)
Accuracy89.3
188
Visual Question AnsweringGQA (test-dev)
Accuracy72.1
184
Visual ReasoningGQA
Accuracy64.54
93
Visual Question AnsweringGQA
GQA Score64.83
85
Visual Question AnsweringGQA (test-std)
Accuracy65.65
68
Object Hallucination ProbingGQA POPE Popular
Accuracy86.07
49
Object Hallucination ProbingGQA POPE Random
Accuracy (GQA POPE)89.93
42
Object Hallucination ProbingGQA Adversarial
Accuracy82.73
40
Visual Question AnsweringGQA
GQA Score63.4
37
Visual Question AnsweringGQA
Accuracy75.77
36
Multi-modal Vision-Language UnderstandingGQA
Accuracy63.4
36
Multi-turn Visual Question AnsweringMT-GQA
Acc165.45
33
Visual Question AnsweringGQA balanced (test-dev)
Accuracy77.4
32
Visual Question AnsweringGQA (val)
Accuracy83.39
32
Visual Question AnsweringGQA v1.0 (test)
Accuracy63.3
31
Refusal Rate EvaluationGQA
Refusal Rate77
30
Visual Question AnsweringGQA
Accuracy65.4
30
Visual Question AnsweringGQA
Accuracy63.9
29
Object Hallucination EvaluationGQA (Random)
Accuracy89.5
28
Visual Question AnsweringGQA v1.2 (test)
GQA Score61.9
28
Visual Question AnsweringGQA
ECE6.09
27
Showing 25 of 94 rows