Design Solutions on Design Solutions
75.26Preference AccuracyDICAI
Evaluation Results
| Method | Links | |
|---|---|---|
| DICAIJudge Model=GPT-4o2026.06 | 75.26 | |
| ICAIJudge Model=GPT-4o2026.06 | 75.2 | |
| CoTJudge Model=GPT-4o2026.06 | 49.25 | |
| AutoRubricJudge Model=GPT-4o2026.06 | 45.05 | |
| CoT-SCJudge Model=GPT-4o2026.06 | 42.29 | |
| ToTJudge Model=GPT-4o2026.06 | 40.2 | |
| Self-RefineJudge Model=GPT-4o2026.06 | 38.61 |