Multi-Document Summarization on WCEP 50 (test)
43ROUGE-1PRIMERA (reported)
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| PRIMERA (reported)Evaluation Setting=100-shot, Maximum Length=502022.08 | 43 | 20.5 | 33.9 | |
| PRIMERA (reproduced)Evaluation Setting=100-shot, Maximum Length=502022.08 | 42.4 | 20.5 | 33.4 | |
| Pegasus (reported)Evaluation Setting=100-shot, Maximum Length=502022.08 | 42.1 | 19.9 | 33 | |
| CentrumEvaluation Setting=100-shot, Maximum Length=502022.08 | 42 | 20.1 | 33 | |
| PRIMERA (reported)Evaluation Setting=10-shot, Maximum Length=502022.08 | 39 | 17.6 | 30.6 | |
| CentrumEvaluation Setting=10-shot, Maximum Length=502022.08 | 38.2 | 17 | 29.5 | |
| PRIMERA (reproduced)Evaluation Setting=10-shot, Maximum Length=502022.08 | 37 | 15.8 | 28.2 | |
| CentrumEvaluation Setting=Zero-shot, Maximum Length=502022.08 | 35.7 | 14.2 | 25.8 | |
| Pegasus (reported)Evaluation Setting=10-shot, Maximum Length=502022.08 | 35.6 | 14.8 | 26.8 | |
| Pegasus (reported)Evaluation Setting=Zero-shot, Maximum Length=502022.08 | 33.2 | 12.7 | 23.8 | |
| PRIMERA (reproduced)Evaluation Setting=Zero-shot, Maximum Length=502022.08 | 32.9 | 12.1 | 23.4 | |
| PRIMERA (reported)Evaluation Setting=Zero-shot, Maximum Length=502022.08 | 28 | 10.3 | 20.9 |