Preference Accuracy on Research Questions
0.8021Preference AccuracyDICAI
Evaluation Results
| Method | Links | |
|---|---|---|
| DICAIJudge Model=GPT-4o2026.06 | 0.8021 | |
| ICAIJudge Model=GPT-4o2026.06 | 0.702 | |
| CoTJudge Model=GPT-4o2026.06 | 0.568 | |
| Self-RefineJudge Model=GPT-4o2026.06 | 0.5379 | |
| ToTJudge Model=GPT-4o2026.06 | 0.5355 | |
| AutoRubricJudge Model=GPT-4o2026.06 | 0.5355 | |
| CoT-SCJudge Model=GPT-4o2026.06 | 0.5078 |