Name
Abstract | ||
|---|---|---|
Knowing which words have been attended to in previous time steps while generating a translation is a rich source of information for predicting what words will be attended to in the future. We improve upon the attention model of Bahdanau et al. (2014) by explicitly modeling the relationship between previous and subsequent attention levels for each word using one recurrent network per input word. This architecture easily captures informative features, such as fertility and regularities in relative distortion. In experiments, we show our parameterization of attention improves translation quality. |
Year | DOI | Venue |
|---|---|---|
2016 | 10.18653/v1/e17-2061 | conference of the european chapter of the association for computational linguistics |
Field | DocType | Volume |
Architecture,Computer science,Machine translation,Attention model,Natural language processing,Artificial intelligence,Distortion,Machine learning | Journal | abs/1607.05108 |
Citations | PageRank | References |
9 | 0.52 | 11 |
Authors | ||
5 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Zichao Yang | 1 | 783 | 26.81 |
| Zhiting Hu | 2 | 758 | 32.20 |
| yuntian deng | 3 | 241 | 14.12 |
| chris dyer | 4 | 5438 | 232.28 |
| Alexander J. Smola | 5 | 19627 | 1967.09 |