Explainability analysis

30 programs randomly sampled from the 1,100 human-rated samples; seed 42.

Score consistency across runs

Three runs per sample · 0–20 scale

ModelIdentical scoresMean score rangeMaximum score range
GPT-6.1 Sol14/30 (46.7%)0.531
DeepSeek V4 Pro3/30 (10.0%)3.7014

Sampled examples


Runs 1 / 2 / 3

Runs 1 / 2 / 3
E01 · Dorn/java/109Dorn17 / 16 / 1616 / 15 / 140.6420.922
E02 · Buse/55Buse13 / 13 / 1212 / 8 / 150.4460.202
E03 · JetBrains/i49JetBrains17 / 17 / 1716 / 13 / 140.3130.246
E04 · JetBrains/e14JetBrains18 / 18 / 1818 / 18 / 170.4780.288
E05 · Dorn/python/97Dorn15 / 15 / 1617 / 15 / 160.6180.947
E06 · Dorn/java/166Dorn16 / 15 / 1516 / 14 / 130.6350.538
E07 · Dorn/cuda/9Dorn15 / 16 / 1616 / 14 / 120.2710.483
E08 · Dorn/cuda/61Dorn13 / 13 / 1312 / 14 / 140.2470.212
E09 · Schnappinger/aoi/artofillusion.view.GLCanvasDrawerSchnappinger15 / 15 / 1514 / 16 / 140.1480.023
E10 · Buse/68Buse18 / 17 / 1717 / 3 / 160.5910.51
E11 · Buse/64Buse16 / 16 / 174 / 14 / 130.4690.596
E12 · Dorn/cuda/73Dorn14 / 14 / 136 / 8 / 40.5210.409
E13 · Dorn/python/88Dorn18 / 17 / 1712 / 2 / 100.860.944
E14 · JetBrains/b39JetBrains18 / 18 / 1818 / 17 / 180.5750.987
E15 · Schnappinger/junit4/junit.framework.TestCaseSchnappinger16 / 16 / 1616 / 16 / 160.7020.804
E16 · Buse/58Buse18 / 18 / 1814 / 16 / 150.4660.101
E17 · Dorn/python/51Dorn16 / 16 / 167 / 18 / 140.5170.735
E18 · Schnappinger/aoi/artofillusion.translators.TextureImageExporterSchnappinger15 / 14 / 1414 / 17 / 130.1170.205
E19 · Dorn/python/91Dorn15 / 14 / 148 / 6 / 120.5510.652
E20 · Schnappinger/argoUML/org.argouml.uml.diagram.deployment.ui.UMLDeploymentDiagramSchnappinger14 / 15 / 1514 / 14 / 140.4650.119
E21 · JetBrains/i56JetBrains18 / 19 / 1918 / 18 / 170.5780.246
E22 · Buse/20Buse19 / 19 / 1918 / 18 / 180.4180.449
E23 · Dorn/java/207Dorn15 / 15 / 1618 / 17 / 160.620.983
E24 · Schnappinger/aoi/artofillusion.view.TexturedVertexShaderSchnappinger15 / 15 / 1516 / 14 / 150.2220.073
E25 · Scalabrio19Scalabrino19 / 19 / 1918 / 19 / 180.670.734
E26 · Dorn/java/199Dorn15 / 15 / 1517 / 18 / 170.6080.819
E27 · Dorn/python/81Dorn17 / 16 / 1714 / 14 / 70.2890.379
E28 · Scalabrio183Scalabrino16 / 17 / 1616 / 14 / 160.5320.455
E29 · Dorn/cuda/71Dorn15 / 15 / 1514 / 12 / 110.2560.187
E30 · Scalabrio83Scalabrino13 / 13 / 138 / 4 / 70.3930.399