Sinkformers: Transformers with Doubly Stochastic Attention. Sander, M. E., Ablin, P., Blondel, M., & Peyré, G. In Proc. AISTATS'22, 2022.
Sinkformers: Transformers with Doubly Stochastic Attention [link]Paper  bibtex   
@inproceedings{2022-sander-sinkformers,
    title = "Sinkformers: Transformers with Doubly Stochastic Attention",
    author = "M. E. Sander and P. Ablin and M. Blondel and G. Peyr{\'e}",
     booktitle = {Proc. AISTATS'22},
     year = "2022",
     url = {https://arxiv.org/abs/2110.11773},
}

Downloads: 0