Learning to Reason: End-to-End Module Networks for Visual Question Answering
Open paper
Benchmark
Model
Rank
Results
visual-dialog-on-visual-dialog-v1-0-test-std
NMN
#4
NDCG (x 100): 58.1
MRR (x 100): 58.8
R@1: 44.15
R@5: 76.88
…
visual-question-answering-on-vqa-v2-test-dev
N2NMN (ResNet-152, policy search)
#40
Accuracy: 64.9
Rank counts only results with a code link.