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

Code Correctness Classification on LiveSQLBench SQLite

0.842AUROC

Functional Sets Confidence

0.5040.591750.67950.76725May 27, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.05
0.842
2026.05
0.84
2026.05
0.818
2026.05
0.782
2026.05
0.778
2026.05
0.775
2026.05
0.75
2026.05
0.749
2026.05
0.749
2026.05
0.745
2026.05
0.745
2026.05
0.743
2026.05
0.739
2026.05
0.738
2026.05
0.737
2026.05
0.726
2026.05
0.721
2026.05
0.712
2026.05
0.706
2026.05
0.705
2026.05
0.702
2026.05
0.695
2026.05
0.694
2026.05
0.687
2026.05
0.687
2026.05
0.683
2026.05
0.679
2026.05
0.678
2026.05
0.678
2026.05
0.674
2026.05
0.673
2026.05
0.672
2026.05
0.665
2026.05
0.662
2026.05
0.658
2026.05
0.657
2026.05
0.654
2026.05
0.647
2026.05
0.646
2026.05
0.641
2026.05
0.632
2026.05
0.629
2026.05
0.628
2026.05
0.623
2026.05
0.616
2026.05
0.602
2026.05
0.596
2026.05
0.596
2026.05
0.558
2026.05
0.554
2026.05
0.546
2026.05
0.535
2026.05
0.529
2026.05
0.528
2026.05
0.517