Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Selective Generation on MMLU

0.947ROC-AUC

TAD

0.418680.555840.6930.83016Aug 20, 2024
Updated 1mo ago

Evaluation Results

MethodLinks
2024.08
0.947
2024.08
0.945
2024.08
0.939
2024.08
0.939
2024.08
0.928
2024.08
0.909
2024.08
0.888
2024.08
0.884
2024.08
0.881
2024.08
0.879
2024.08
0.874
2024.08
0.855
2024.08
0.847
2024.08
0.846
2024.08
0.846
2024.08
0.841
2024.08
0.84
2024.08
0.838
2024.08
0.83
2024.08
0.828
2024.08
0.82
2024.08
0.816
2024.08
0.813
2024.08
0.806
2024.08
0.795
2024.08
0.772
2024.08
0.771
2024.08
0.768
2024.08
0.758
2024.08
0.745
2024.08
0.744
2024.08
0.741
2024.08
0.739
2024.08
0.738
2024.08
0.735
2024.08
0.733
2024.08
0.725
2024.08
0.724
2024.08
0.718
2024.08
0.709
2024.08
0.709
2024.08
0.697
2024.08
0.697
2024.08
0.673
2024.08
0.673
2024.08
0.663
2024.08
0.661
2024.08
0.654
2024.08
0.642
2024.08
0.64
2024.08
0.637
2024.08
0.637
2024.08
0.633
2024.08
0.631
2024.08
0.63
2024.08
0.63
2024.08
0.627
2024.08
0.625
2024.08
0.622
2024.08
0.621
2024.08
0.619
2024.08
0.604
2024.08
0.603
2024.08
0.6
2024.08
0.59
2024.08
0.439