Datasets:

ccdv
/

WCEP-10

Tasks:

Text2Text Generation

Languages:

Multilinguality:

Size Categories:

ArXiv:

Tags:

conditional-text-generation

Dataset card Viewer Files Files and versions Community

ccdv commited on May 21, 2022

Commit

0589d4a

•

1 Parent(s): ce19fd0

readme

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -22,7 +22,7 @@ This dataset is compatible with the [`run_summarization.py`](https://github.com/
 # Configs
 4 possibles configs:
-- `roberta` will concatenate documents with "</s>" (default)
 - `newline` will concatenate documents with "\n"
 - `bert` will concatenate documents with "[SEP]"
 - `list` will return the list of documents instead of a string

 # Configs
 4 possibles configs:
+- `roberta` will concatenate documents with "\</s\>" (default)
 - `newline` will concatenate documents with "\n"
 - `bert` will concatenate documents with "[SEP]"
 - `list` will return the list of documents instead of a string