-former: Infinite Memory TransformerAnnual Meeting of the Association for Computational Linguistics (ACL), 2021 |
Sparse Continuous Distributions and Fenchel-Young Losses André F. T. Martins Marcos Vinícius Treviso António Farinhas P. Aguiar Mário A. T. Figueiredo Mathieu Blondel Vlad Niculae |