Unsupervised Kg To Text Generation On Genwiki 1
Metrics
BLEU
CIDEr
METEOR
ROUGE-L
Results
Performance results of various models on this benchmark
Model Name | BLEU | CIDEr | METEOR | ROUGE-L | Paper Title | Repository |
|---|---|---|---|---|---|---|
| CycleGT_Warm | 40.47 | 3.48 | 34.84 | 63.40 | GenWiki: A Dataset of 1.3 Million Content-Sharing Text and Graphs for Unsupervised Graph-to-Text Generation | |
| CycleGT_Base | 41.29 | 3.53 | 35.39 | 63.73 | GenWiki: A Dataset of 1.3 Million Content-Sharing Text and Graphs for Unsupervised Graph-to-Text Generation | |
| DirectTransfer | 13.89 | 1.26 | 25.76 | 39.75 | GenWiki: A Dataset of 1.3 Million Content-Sharing Text and Graphs for Unsupervised Graph-to-Text Generation | |
| Rule-Based | 13.45 | 1.26 | 30.72 | 40.93 | GenWiki: A Dataset of 1.3 Million Content-Sharing Text and Graphs for Unsupervised Graph-to-Text Generation | |
| NoisySupervised | 35.03 | 2.63 | 33.45 | 58.14 | GenWiki: A Dataset of 1.3 Million Content-Sharing Text and Graphs for Unsupervised Graph-to-Text Generation |
0 of 5 row(s) selected.