I don't understand how this helps in improving performance. Can you elaborate?
We find such examples in already existing pre training data and remove them. Do you not think it will work?
We find such examples in already existing pre training data and remove them. Do you not think it will work?