all
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
All Posts and Comment Mean
68.51Mean Score
42
ALL Avg
86.87CQ Score
25
All Average
60Accuracy
20
All weighted average (test)
0.068BLEU Score
14
All weighted average (test)
10.28BLEU Score
14
ALL
0.913PLCC
12
ALL Average of two language pairs in four directions wmt22-comet-da
85.8COMET
12
All 121 classes v1 (test)
90.49DSC
10
All-50K (test)
8.8HR@1
9
ALL (test)
82F1 Score
8
ALL (Hold-Out)
93AUC
7
All 169 Graphs (5-fold stratified CV)
75.9Accuracy (Test)
6
ALL FULL
80.4Precision
5
All (Avg.)
65.5Base Score
5
All Default, Shoppers, Adult
13.47Memory Ratio Improvement (%)
4
all (test)
66.784Line Accuracy
4
All 31 shapes
2.08Mean Chamfer Distance
3
All Average
40.3Accuracy
3
ALL FULL (test)
80.4Precision
3
All MVTec-AD, VisA, MPDD, BTAD combined
95.4I-AUROC
2
All 6 tabular
0.49ΔNLL (%)
1
All Aggregated (UK-based participants)
5.2Final Accuracy
1