Preference Reconstruction on Design Solutions
80.21Preference AccuracyDICAI
Evaluation results
| Method | Links | |
|---|---|---|
| DICAIJudge LLM=GPT-52026.06 | 80.21 | |
| ICAIJudge LLM=GPT-52026.06 | 79.8 | |
| CoTJudge LLM=GPT-52026.06 | 46.27 | |
| AutoRubricJudge LLM=GPT-52026.06 | 45.88 | |
| Self-RefineJudge LLM=GPT-52026.06 | 45.05 | |
| ToTJudge LLM=GPT-52026.06 | 43.56 | |
| CoT-SCJudge LLM=GPT-52026.06 | 43.5 |