Datasets:

Languages:
English
Multilinguality:
monolingual
Size Categories:
n<1K
Language Creators:
found
Annotations Creators:
found
Source Datasets:
original
Tags:
License:

Possibly Tagged under Wrong "Tasks"

#4
by jjjone - opened

The "Tasks" tags list this as "Text2Text Generation" and "Text Classification". How does this fall under "Text2Text Generation"? Seems to me it should only be TC. When I think of T2TG, I think of text inputs and text targets. Not text inputs and classification targets. Please consider removing the "Text2Text Generation" tag to help users who use the Tasks filter.

Datasets Maintainers org

Hi @jjjone , thanks for pointing this unclear issue.

Indeed, this dataset provides 189 texts, each in three reading level versions (567 in total): elementary, intermediate and advance.

So, the texts put in parallel, can also be used for automatic text simplification (Text2Text Generation).

I agree with you that this is not evident just looking at the current format of the dataset. Maybe, it would be interesting to implement two configurations of this dataset:

  • one, as it is now, for text classification
  • and other, with the texts in parallel, for text simplification

Sign up or log in to comment