Multiple
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Multiple Aggregate
69.8Average Score
18
Multiple (ImageNet-1K, COCO)
-11.8Mean Delta
13
Multiple (test)
1.9AE Rank (T=1)
6
Multiple TriviaQA, HotpotQA, CSQA
72.9Average AUROC
4
Multiple
78.51Score
3
Multiple Distributional Constraint
0.95Ctrl
3
Multiple Tabular, Text, Image (test)
—Primary metric
0