Example: unclear how to add repeated datasets

[dgarijo]

It should be repeated, but we may state that adding the same identifier is enough

If for evaluated on (ind:evaluatedOn) and tested on(ind:testedOn), the
dataset used is the same as trained on (ind:trainedOn), for example
masked language modeling where we mask words, do we just copy the same
dataset?, I wrote # no additional dataset were used for validation for
both evaluated on (ind:evaluatedOn) and tested on(ind:testedOn).