Guided Alignment Training for Topic-Aware Neural Machine Translation | Synced
The author uses a hard-attention (Viterbi alignment matrix) of the SMT (Statistical Machine Translation) to help learn a better alignment matrix in NMT during the training.
Source: Synced | AI Technology & Industry Review
The author uses a hard-attention (Viterbi alignment matrix) of the SMT (Statistical Machine Translation) to help learn a better alignment matrix in NMT during the training.