| Bi-SimCut: A Simple Strategy for Boosting Neural Machine Translation | ✓ Link | 35.15 | Bi-SimCut | 2022-06-06 |
| BERT, mBERT, or BiBERT? A Study on Contextualized Embeddings for Neural Machine Translation | ✓ Link | 34.94 | BiBERT | 2021-09-09 |
| Bi-SimCut: A Simple Strategy for Boosting Neural Machine Translation | ✓ Link | 34.86 | SimCut | 2022-06-06 |
| Mega: Moving Average Equipped Gated Attention | ✓ Link | 33.12 | Mega | 2022-09-21 |
| Incorporating a Local Translation Mechanism into Non-autoregressive Translation | ✓ Link | 32.04 | CMLM+LAT+4 iterations | 2020-11-12 |
| Wide-minima Density Hypothesis and the Explore-Exploit Learning Rate Schedule | ✓ Link | 31.9 | MAT+Knee | 2020-03-09 |
| Non-Autoregressive Translation by Learning Target Categorical Codes | ✓ Link | 30.75 | CNAT | 2021-03-21 |
| Incorporating a Local Translation Mechanism into Non-autoregressive Translation | ✓ Link | 29.91 | CMLM+LAT+1 iterations | 2020-11-12 |
| FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow | ✓ Link | 28.29 | FlowSeq-large (NPD n = 30) | 2019-09-05 |
| FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow | ✓ Link | 27.71 | FlowSeq-large (NPD n = 15) | 2019-09-05 |
| FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow | ✓ Link | 27.16 | FlowSeq-large (IWD n=15) | 2019-09-05 |
| Deterministic Non-Autoregressive Neural Sequence Modeling by Iterative Refinement | ✓ Link | 25.43 | Denoising autoencoders (non-autoregressive) | 2018-02-19 |
| FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow | ✓ Link | 25.4 | FlowSeq-large | 2019-09-05 |
| FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow | ✓ Link | 23.36 | FlowSeq-base | 2019-09-05 |
| Non-Autoregressive Neural Machine Translation | ✓ Link | 23.20 | NAT +FT + NPD | 2017-11-07 |
| Unsupervised Statistical Machine Translation | ✓ Link | 17.43 | SMT + iterative backtranslation (unsupervised) | 2018-09-04 |