You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In the MultiRC dataset, the spans are whole sentences. Therefore, span_start is suffice to represent a span.
i don not understand what the task is, what i understand is , given a paragraph contain multi-sentences, a question and some answer options, then we should choose the right options. why use bidaf to predict the sentences no predict option?
how to get span_end for multibidaf?
The text was updated successfully, but these errors were encountered: