Optimizing Pre-Training Data Mixtures with Mixtures of Data Expert Models.

Lior Belenki, Alekh Agarwal, Tianze Shi, Kristina Toutanova

Browse the full ACL paper archive.