1 code implementation • 18 Oct 2021 • Hareesh Thuruthipilly, Adam Zadrozny, Agnieszka Pollo, Marek Biesiada
The performances of self-attention-based encoder models and CNNs participating in the challenge are compared.