Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Classification on ColorMNIST (OOD)
Loading...
73.8
Accuracy
ERM
-2.952
16.974
36.9
56.826
Nov 24, 2025
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
ERM
shortcut severity (ρ)=0.5
2025.11
73.8
SITAR
shortcut severity (ρ)=0.9
2025.11
71.5
SITAR
shortcut severity (ρ)=1.0
2025.11
70.3
SITAR
shortcut severity (ρ)=0.7
2025.11
70.2
SITAR
shortcut severity (ρ)=0.5
2025.11
69.9
LfF
shortcut severity (ρ)=0.9
2025.11
69.6
ERM
shortcut severity (ρ)=0.7
2025.11
66.7
LfF
shortcut severity (ρ)=0.7
2025.11
64.3
JTT
shortcut severity (ρ)=0.9
2025.11
61.2
LfF
shortcut severity (ρ)=0.5
2025.11
58.5
JTT
shortcut severity (ρ)=0.5
2025.11
56.3
JTT
shortcut severity (ρ)=0.7
2025.11
53.6
ERM
shortcut severity (ρ)=0.9
2025.11
27.3
ERM
shortcut severity (ρ)=1.0
2025.11
0
JTT
shortcut severity (ρ)=1.0
2025.11
0
LfF
shortcut severity (ρ)=1.0
2025.11
0
Feedback
Search any
task
Search any
task