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.- Anthology ID:
- D18-1149
- Volume:
- Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing
- Month:
- October-November
- Year:
- 2018
- Address:
- Brussels, Belgium
- Editors:
- Ellen Riloff, David Chiang, Julia Hockenmaier, Jun’ichi Tsujii
- Venue:
- EMNLP
- SIG:
- SIGDAT
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 1173–1182
- Language:
- URL:
- https://aclanthology.org/D18-1149
- DOI:
- 10.18653/v1/D18-1149
- Cite (ACL):
- Jason Lee, Elman Mansimov, and Kyunghyun Cho. 2018. Deterministic Non-Autoregressive Neural Sequence Modeling by Iterative Refinement. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 1173–1182, Brussels, Belgium. Association for Computational Linguistics.
- Cite (Informal):
- Deterministic Non-Autoregressive Neural Sequence Modeling by Iterative Refinement (Lee et al., EMNLP 2018)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-2/D18-1149.pdf
- Code
- nyu-dl/dl4mt-nonauto + additional community code
- Data
- MS COCO, WMT 2014, WMT 2016, WMT 2016 News