Hypothesis Generation
77.22Preference AccuracyDICAI
Evaluation Results
| Method | Links | |
|---|---|---|
| DICAIJudge Model=GPT-4o2026.06 | 77.22 | |
| ICAIJudge Model=GPT-4o2026.06 | 64.6 | |
| CoTJudge Model=GPT-4o2026.06 | 56.1 | |
| ToTJudge Model=GPT-4o2026.06 | 55.42 | |
| CoT-SCJudge Model=GPT-4o2026.06 | 54.96 | |
| AutoRubricJudge Model=GPT-4o2026.06 | 54.92 | |
| Self-RefineJudge Model=GPT-4o2026.06 | 52.56 |