Conversation Evaluation on CRSArena-Eval Dial-level
0.913Pearson rFACE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| FACE2026.06 | 0.913 | 0.837 | |
| G-Evaltype=reference-free2026.06 | 0.893 | 0.83 | |
| LLMDirecttype=reference-free2026.06 | 0.872 | 0.799 | |
| Distinct-3n-gram=32026.06 | 0.665 | 0.78 | |
| Distinct-4n-gram=42026.06 | 0.609 | 0.76 | |
| iEvaLMattrmode=attr2026.06 | 0.553 | 0.553 | |
| iEvaLMfreemode=free2026.06 | 0.232 | 0.232 | |
| R@10type=reference-based2026.06 | -0.111 | 0.071 | |
| R@1type=reference-based2026.06 | -0.12 | 0.081 |