Paper Info
Title
Deterministic Non-Autoregressive Neural Sequence Modeling by Iterative Refinement.
Abstract
We propose a conditional non-autoregressive neural sequence model based on iterative refinement. The proposed model is designed based on the principles of latent variable models and denoising autoencoders, and is generally applicable to any sequence generation task. We extensively evaluate the proposed model on machine translation (En-De and En-Ro) and image caption generation, and observe that it significantly speeds up decoding while maintaining the generation quality comparable to the autoregressive counterpart.
Year
DOI
Venue
2018
10.18653/v1/d18-1149
EMNLP
DocType
Volume
Citations 
Conference
abs/1802.06901
11
PageRank 
References 
Authors
0.50
23
3
Name
Order
Citations
PageRank
Lee, Jason D.171148.29
elman mansimov238713.74
Kyunghyun Cho36803316.85