LXMERT: Learning Cross-Modality Encoder Representations from Transformers.

Hao Tan, Mohit Bansal

Browse the full EMNLP paper archive.