Code Generation on HumanEval (AUROC, AUPRC, ACC*) benchmark leaderboard · SOTA2 Research

Code Generation on HumanEval (AUROC, AUPRC, ACC*)

80.9ACC*TokUR

Evaluation Results

Method	Links
TokUR2025.05		80.9	63.15	81.59
TokUR2025.05		80.9	60.5	81.75
Self-Certainty2025.05		80.49	61.48	80.27
DeepConf2025.05		80.49	63.6	82.75
TokUR2025.05		80.49	60.64	80.75
PE2025.05		78.05	60.52	81.45
LL2025.05		75.61	56.42	79.48
CoT2025.05		74.39	—	—