OOD
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
OOD
82.19SSIM
4
OOD-3 Layout
23Success Rate (OOD-3 All)
4
OOD-2 Layout
12.88OOD-2 Success Rate (All)
4
OOD Average
60.5Win Rate
4
OOD Table S2 (test)
40.38ROUGE-L
4
OOD Table S1 (test)
80.2Accuracy
4
OOD Multimodal
64Average Sample Size (N)
3
OOD Text
64Average N
3
ood-large 285 polygons, n from 600 to 2250
96.3Mean Coverage
3
ood 2081 polygons, n up to 1000 (out-of-distribution)
0.995Mean Coverage
3
OOD Out-of-Distribution (test)
1.07ERG (All\Vt)
3
OOD 29 geographically disjoint scenes (test)
31.68PSNR
3
OOD (Out-of-Distribution)
0.052ECE
3
OOD Physical Track (a) 1/10th scale (test)
9.56Lap Time (s)
3
OOD Average-5
0.919F-Score
2
OOD WindowR
89.1F-Score
2
OOD Suitcase
90F-Score
2
OOD Box
92.7F-Score
2
OOD Lighter
92.3F-Score
2
OOD Car
95.5F-Score
2
OOD Average-10
91.6F-Score
1
OOD Skateboard
94.3F-Score
1
Overall OOD (test)
97.4Accuracy
1
OOD deployment set
—Primary metric
0