Datasets:
Tasks:
Text Generation
License:
The Dataset Viewer has been disabled on this dataset.
TEKGEN is a training corpus utilized by Google for their KELM paper. In this repository, I have uploaded the knowledge graphs and the original sentences used to create these graphs for the KELM dataset. Note that this dataset is not the KELM dataset itself but rather the training corpus used for the REALM model.
- Downloads last month
- 13