Text Simplification On Asset
評価指標
BLEU
SARI (EASSEu003e=0.2.1)
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 | BLEU | SARI (EASSEu003e=0.2.1) |
|---|---|---|
| controllable-sentence-simplification | 75.99* | 40.13 |
| the-gem-benchmark-natural-language-generation | - | - |
| the-gem-benchmark-natural-language-generation | - | - |
| sentence-simplification-by-monolingual | 79.39* | 34.63 |
| control-prefixes-for-text-generation | - | 43.58 |
| text-simplification-by-tagging | - | 43.21 |
| sentence-simplification-with-deep | 86.39* | 36.59 |
| metric-based-in-context-learning-a-case-study | 73.92 | 47.94 |
| unsupervised-neural-text-simplification | 76.14* | 35.19 |
| multilingual-unsupervised-sentence | - | 42.65 |
| multilingual-unsupervised-sentence | 72.98 | 44.15 |
| integrating-transformer-and-paraphrase-rules | 71.44* | 38.67 |