Text Summarization On Samsum Corpus
評価指標
ROUGE-1
ROUGE-L
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
比較表
| モデル名 | ROUGE-1 | ROUGE-L |
|---|---|---|
| salient-information-prompting-to-steer | 44.1 | 33.9 |
| omnivec2-a-novel-transformer-based-network | 59.1 | 63.7 |
| topic-aware-contrastive-learning-for | 54.3 | 45.2 |
| momentum-calibration-for-text-generation | 55.13 | 50.88 |
| mind-the-gap-injecting-commonsense-knowledge | 53.73 | 49.5 |
| hierarchical-learning-for-generation-with | 53.01 | - |
| instructive-dialogue-summarization-with-query | 55.3 | 46.7 |
| hierarchical-learning-for-generation-with | - | 48.84 |
| hierarchical-prompting-taxonomy-a-universal | - | 30 |
| crispo-multi-aspect-critique-suggestion | 47.2 | 38.2 |
| calibrating-sequence-likelihood-improves | 54.37 | 45.89 |
| topic-aware-contrastive-learning-for | 52.6 | 42.1 |