Learning to Defer
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
28.1Test True Risk
6
Mar 17, 2026
88SAC
6
Mar 4, 2026
89SAC
6
Mar 4, 2026
86SAC
6
Mar 4, 2026
90Balanced Accuracy
4
May 5, 2026
72.1Balanced Accuracy
4
May 5, 2026
84.1Balanced Accuracy
4
May 5, 2026
85.3Balanced Accuracy
4
May 5, 2026
12.59Error
4
Feb 26, 2026
8.06Error
4
Feb 26, 2026
64.56Accuracy
3
May 20, 2026
83.39Classification Accuracy
2
May 21, 2026
76.59Accuracy
2
May 20, 2026
87.74Accuracy
2
May 20, 2026
45.64Accuracy
2
May 20, 2026
61.41Accuracy
2
May 20, 2026
87.31Accuracy
2
May 20, 2026