fix task_ids

README.md

@@ -15,10 +15,11 @@ source_datasets:
 - original
 task_categories:
 - other
-task_ids:
-- other-other-data-to-text-generation
+task_ids: []
 paperswithcode_id: kelm
 pretty_name: Corpus for Knowledge-Enhanced Language Model Pre-training (KELM)
+tags:
+- data-to-text-generation
 ---
 
 # Dataset Card for Corpus for Knowledge-Enhanced Language Model Pre-training (KELM)
@@ -161,4 +162,4 @@ This dataset has been released under the [CC BY-SA 2.0 license](https://creative
 
 ### Contributions
 
-Thanks to [@joeddav](https://github.com/joeddav) for adding this dataset.
+Thanks to [@joeddav](https://github.com/joeddav) for adding this dataset.