Metaphors on Metaphors (Preference accuracy)
74.01Preference AccuracyDICAI
Evaluation Results
| Method | Links | |
|---|---|---|
| DICAIJudge Model=GPT-4o2026.06 | 74.01 | |
| ICAIJudge Model=GPT-4o2026.06 | 71.4 | |
| CoT-SCJudge Model=GPT-4o2026.06 | 60 | |
| ToTJudge Model=GPT-4o2026.06 | 60 | |
| Self-RefineJudge Model=GPT-4o2026.06 | 60 | |
| AutoRubricJudge Model=GPT-4o2026.06 | 60 | |
| CoTJudge Model=GPT-4o2026.06 | 40 |