transformer neural-machine-translation average-attention-network aan speech-translation depth-scaled-initialization deep-transformer l0drop adaptive-feature-selection massively-multilingual-translation opus-100 fast-bidirectional-decoder

average-attention-network transformer-aan aan

natural-language-processing python3 pytorch transformer seq2seq ensemble beam-search neural-machine-translation multi-gpu attention-is-all-you-need multilingual-nmt average-attention-network dynamic-sentence-sampling robust-neural-machine-translation average-models optimizers sentential-context context-aware-nmt relative-position dynamic-batch-size