Summarization (Relevance) on DialSummEval
0.69Kendall's TauJury-on-Demand
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Jury-on-DemandJury Configuration=Dynamic2025.12 | 0.69 | 0.05 | |
| Static JuryJury Configuration=Weighted-Tau2025.12 | 0.66 | — | |
| Static JuryJury Configuration=Average-All2025.12 | 0.65 | 0.02 | |
| Static JuryJury Configuration=Weighted-Regression2025.12 | 0.65 | 0.02 | |
| Static JuryJury Configuration=Average-TopK2025.12 | 0.63 | 0.02 |