the QA cases comparing pre- and post- edit in paper cannot be found from the dataset, where can we find the cases, such as “Who design the Burj Khalifa?”
LeiyanGithub opened this issue · 1 comments
LeiyanGithub commented
the QA cases comparing pre- and post- edit in paper cannot be found from the dataset, where can we find the cases, such as “Who design the Burj Khalifa?”
eric-mitchell commented
These prompts were manually constructed (typed in by hand) and evaluated with a pre-trained MEND model.