Papers with EM-like process
Almost Free Semantic Draft for Neural Machine Translation (2021.naacl-main)
Copied to clipboard
| Challenge: | Empirical experiments show that the presented method can achieve competitive performance in common language pairs with a clear advantage in inference efficiency. |
| Approach: | They propose a method to sample and consider a semantic draft as global information from semantic space for decoding with almost free of cost. |
| Outcome: | Empirical results show that the proposed method can achieve competitive performance in common language pairs with a clear advantage in inference efficiency. |