Conversation Evaluation on CRSArena-Eval Turn-level
0.93Pearson Correlation (r)FACE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| FACE2026.06 | 0.93 | 0.842 | |
| LLMDirecttype=reference-free2026.06 | 0.86 | 0.822 | |
| G-Evaltype=reference-free2026.06 | 0.74 | 0.84 | |
| Distinct-3n-gram=32026.06 | 0.716 | 0.841 | |
| Distinct-4n-gram=42026.06 | 0.654 | 0.8 | |
| iEvaLMattrmode=attr2026.06 | 0.527 | 0.527 | |
| iEvaLMfreemode=free2026.06 | 0.191 | 0.191 | |
| R@10type=reference-based2026.06 | -0.192 | 0.048 | |
| R@1type=reference-based2026.06 | -0.197 | 0.06 |