-
Update README.md
-
Delete dataset_unrelevant.json
-
Delete dataset_mistake.json
-
Update README.md
-
split the mistake short
-
Update README.md
-
Update README.md
-
Update README.md
-
fix chore: fix config
-
split the files
-
Upload dataset_unrelevant.json
Add remove unrelevant topic sentences for 1000 data of 350 word article version -
Add remove mistake datasets
Add two datasets for 100 topics
remove mistake for 350 words article version
remove mistake short for 150 words short article version -
Delete dataset_mistake.json
-
Add Description of the dataset
-
Upload dataset_mistake.json
-
Update README.md
It is a dataset for training remove mistake sentences in a content -
initial commit