Multi-document summarization on Multi-News (test)
149Non-RedundancyHuman
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| HumanEvaluation Protocol=Pairwise Best-Worst Scaling, Summary Type=Gold2019.06 | 149 | 150 | 150 | — | — | — | |
| CopyTransformerEvaluation Protocol=Pairwise Best-Worst Scaling2019.06 | 107 | 99 | 100 | — | — | — | |
| Hi-MAPEvaluation Protocol=Pairwise Best-Worst Scaling2019.06 | 100 | 85 | 75 | — | — | — | |
| PG-MMREvaluation Protocol=Pairwise Best-Worst Scaling2019.06 | 45 | 95 | 70 | — | — | — | |
| MoA2026.06 | 4.67 | — | — | 4.56 | 4.55 | 4.47 | |
| Centrum2026.06 | 3.35 | — | — | 3.57 | 3.67 | 3.37 | |
| PRIMERA2026.06 | 3.15 | — | — | 2.75 | 3.1 | 2.75 |