Neural Machine Translation by Jointly Learning to Align and Translate
Bahdanau et al., 2015, Neural Machine Translation by Jointly Learning to Align and Translate (3rd International Conference on Learning Representations (ICLR 2015))
Subscriber deck — free during early access
1 cards in “7 Conclusion” · whole deck
7 Conclusion
- What challenge did Bahdanau et al. leave for future work?Better handling of unknown and rare words, needed to match state-of-the-art systems in all contexts; their models used a 30,000-word shortlist with everything else mapped to [UNK].
- Bahdanau et al., 2015, Neural Machine Translation by Jointly Learning to Align and Translate (3rd International Conference on Learning Representations (ICLR 2015)), §7
Source
- Dzmitry Bahdanau, Kyunghyun Cho and Yoshua Bengio. 2015. Neural Machine Translation by Jointly Learning to Align and Translate. In 3rd International Conference on Learning Representations (ICLR 2015). https://arxiv.org/abs/1409.0473