Symmetric Dot-Product Attention for Efficient Training of BERT Language Models.

Martin Courtois, Malte Ostendorff, Leonhard Hennig, Georg Rehm

Browse the full ACL paper archive.