Адаптация домена без обучения

AbstractThis paper describes LIMSI’s submissions to the translation shared tasks at WMT’20. This year we have focused our efforts on the biomedical translation task, developing a resource-heavy system for the …...

AbstractA challenge in the Dialogue State Tracking (DST) field is adapting models to new domains without using any supervised data — zero-shot domain adaptation. Parameter-Efficient Transfer Learning (PETL) has the …...