Human Correlation Analysis on HAMLETJudge (val)
0.792CP CorrelationHamletJudge
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| HamletJudge2025.07 | 0.792 | 0.807 | 0.773 | 0.791 | |
| Gemini-2.5-Pro2025.07 | 0.72 | 0.684 | 0.701 | 0.702 | |
| Claude-4-Sonnet2025.07 | 0.698 | 0.783 | 0.804 | 0.762 | |
| GPT-4.12025.07 | 0.675 | 0.593 | 0.622 | 0.63 |