Data To Text Generation On Xalign
Metriken
BLEU4
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
| Paper Title | ||
|---|---|---|
| Fact-aware embedding with mT5 | 29.27 | XF2T: Cross-lingual Fact-to-Text Generation for Low-Resource Languages |
| Bi-lingual mT5 | 25.88 | XF2T: Cross-lingual Fact-to-Text Generation for Low-Resource Languages |
| mT5 | 25 | XAlign: Cross-lingual Fact-to-Text Alignment and Generation for Low-Resource Languages |
| Vanilla Transformer | 19.9 | XAlign: Cross-lingual Fact-to-Text Alignment and Generation for Low-Resource Languages |
| Translate-Output mT5 | 18.91 | XF2T: Cross-lingual Fact-to-Text Generation for Low-Resource Languages |
| Graph Attention Network Encoder +Transformer Decoder | 18.3 | XAlign: Cross-lingual Fact-to-Text Alignment and Generation for Low-Resource Languages |
0 of 6 row(s) selected.