OOD Generalization
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
83.3ID Accuracy
24
May 27, 2026
58.36Accuracy
20
Feb 27, 2026
80.2OOD Accuracy
18
May 27, 2026
83.1ARC-C Score
13
Jun 30, 2026
0.3Relative L2 Error
5
Mar 9, 2026
0.048Relative L2 Error
5
Mar 9, 2026
0.008Relative L2 Error
5
Mar 9, 2026
0.9Relative L2 Error
5
Mar 9, 2026
0.068Latent MSE
3
May 19, 2026