On the Strengths of Cross-Attention in Pretrained Transformers for Machine Translation. Gheini, M., Ren, X., & May, J. 2021.
bibtex   
@misc{gheini2021strengths,
      title={On the Strengths of Cross-Attention in Pretrained Transformers for Machine Translation}, 
      author={Mozhdeh Gheini and Xiang Ren and Jonathan May},
      year={2021},
      eprint={2104.08771},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

Downloads: 0