Column Type Annotation On Viznet Sato Full
Metrics
Macro-F1
Weighted-F1
Results
Performance results of various models on this benchmark
Model Name | Macro-F1 | Weighted-F1 | Paper Title | Repository |
|---|---|---|---|---|
| Sato | 75.6 | 90.2 | Sato: Contextual Semantic Type Detection in Tables | |
| DODUO | 84.6 | - | Annotating Columns with Pre-trained Language Models | |
| Watchog | 85.63 | - | Watchog: A Light-weight Contrastive Learning based Framework for Column Annotation | - |
| TaBERT | - | 97.2 | TABBIE: Pretrained Representations of Tabular Data |
0 of 4 row(s) selected.