Uncertainty Estimation on HumanEval-X Java (test)
0.69AUROCTop-5 Entropy
Evaluation Results
| Method | Links | |
|---|---|---|
| Top-5 EntropyModel=Qwen3-Coder-30B2026.06 | 0.69 | |
| Consistency (BLEU)Model=Qwen3-Coder-30B2026.06 | 0.682 | |
| Max EntropyModel=Qwen3-Coder-30B2026.06 | 0.679 | |
| Consistency (VR)Model=Qwen3-Coder-30B2026.06 | 0.665 | |
| Mean EntropyModel=Qwen3-Coder-30B2026.06 | 0.624 | |
| Mean ProbabilityModel=Qwen3-Coder-30B2026.06 | 0.502 |