System Ranking Correlation on CRSArena-Eval (All)
0.918Pearson Correlation (r)FACE
Evaluation results
| Method | Links | ||
|---|---|---|---|
| FACE2025.05 | 0.918 | 0.838 | |
| LLMDirect2025.05 | 0.868 | 0.806 | |
| G-Eval2025.05 | 0.85 | 0.833 | |
| Distinct-32025.05 | 0.68 | 0.798 | |
| Distinct-42025.05 | 0.622 | 0.771 | |
| iEvaLMattrUser Simulation=Attribute-based2025.05 | 0.575 | 0.517 | |
| iEvaLMfreeUser Simulation=LLM-based free-form2025.05 | 0.257 | 0.184 | |
| R@102025.05 | -0.134 | 0.064 | |
| R@12025.05 | -0.142 | 0.075 |