Data To Text Generation On Webnlg
評価指標
BLEU
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 |
|---|---|
| control-prefixes-for-text-generation | 67.15 |
| control-prefixes-for-text-generation | 67.32 |
| tricy-trigger-guided-data-to-text-generation-1 | 64.73 |
| factspotter-evaluating-the-factual | 67.04 |
| structural-information-preserving-for-graph-1 | 62.89 |
| step-by-step-separating-planning-from | 47.4 |
| tricy-trigger-guided-data-to-text-generation-1 | 64.08 |
| neural-data-to-text-generation-a-comparison | 57.20 |
| investigating-pretrained-language-models-for | 65.05 |
| modeling-global-and-local-node-contexts-for | 63.69 |
| factspotter-evaluating-the-factual | 67.08 |
| htlm-hyper-text-pre-training-and-prompting-of | 65.4 |
| deep-graph-convolutional-encoders-for | 55.9 |
| text-to-text-pre-training-for-data-to-text | 64.7 |
| stage-wise-fine-tuning-for-graph-to-text | 66.07 |
| textbox-2-0-a-text-generation-library-with | - |
| modeling-graph-structure-via-relative | 61.15 |
| gtr-lstm-a-triple-encoder-for-sentence | 58.6 |