Multilingual Non-Autoregressive Machine Translation without Knowledge Distillation
Not on the map. Why?
Abstract (arXiv)
Multilingual neural machine translation (MNMT) aims at using one single model for multiple translation directions. Recent work applies non-autoregressive Transformers to improve the efficiency of MNMT, but requires expensive knowledge distillation (KD) processes. To this end, we propose an M-DAT approach to non-autoregressive multilingual machine translation. Our system leverages the recent advance of the directed acyclic Transformer (DAT), which does not require KD. We further propose a pivot back-translation (PivotBT) approach to improve the generalization to unseen translation directions. Experiments show that our M-DAT achieves state-of-the-art performance in non-autoregressive MNMT.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 27 of its 32 references that are in the map. First posted on arXiv 6 Feb 2025.
Builds on
27 of 32 references.
References · 27
- 2017Attention Is All You Needcited by 59,932
- 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,530
- 2018A Call for Clarity in Reporting BLEU Scorescited by 1,790
- 2015Improving Neural Machine Translation Models with Monolingual Datacited by 784
- 2016Sequence-Level Knowledge Distillationcited by 761
- 2016Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translationcited by 467
- 2017Non-Autoregressive Neural Machine Translationcited by 457
- 2019Simple, Scalable Adaptation for Neural Machine Translationcited by 239
- 2019Insertion Transformer: Flexible Sequence Generation via Insertion Operationscited by 123
- 2019Understanding Knowledge Distillation in Non-autoregressive Machine Translationcited by 121
- 2016Multi-Way, Multilingual Neural Machine Translation with a Shared Attention Mechanismcited by 117
- 2020Deep Encoder, Shallow Decoder: Reevaluating Non-autoregressive Machine Translationcited by 104
15 more
- 2020Glancing Transformer for Non-Autoregressive Neural Machine Translationcited by 103
- 2020Fully Non-autoregressive Neural Machine Translation: Tricks of the Tradecited by 99
- 2019Improved Zero-shot Neural Machine Translation via Ignoring Spurious Correlationscited by 61
- 2019The Missing Ingredient in Zero-Shot Neural Machine Translationcited by 59
- 2022Directed Acyclic Transformer for Non-Autoregressive Machine Translationcited by 44
- 2021Non-Autoregressive Translation with Layer-Wise Prediction and Deep Supervisioncited by 40
- 2021Non-Autoregressive Text Generation with Pre-trained Language Modelscited by 36
- 2020Improving Zero-Shot Translation by Disentangling Positional Informationcited by 35
- 2021Rejuvenating Low-Frequency Words: Making the Most of Parallel Data in Non-Autoregressive Translationcited by 34
- 2020BANG: Bridging Autoregressive and Non-autoregressive Generation with Large Scale Pretrainingcited by 33
- 2016Neural Machine Translation with Pivot Languagescited by 25
- 2017A Teacher-Student Framework for Zero-Resource Neural Machine Translationcited by 22
- 2021Thinking Clearly, Talking Fast: Concept-Guided Non-Autoregressive Generation for Open-Domain Dialogue Systemscited by 17
- 2022A Character-Level Length-Control Algorithm for Non-Autoregressive Sentence Summarizationcited by 13
- 2022Learning Non-Autoregressive Models from Search for Unsupervised Sentence Summarizationcited by 12
Built on by
Citing papers per year: 2025: 1.
1 paper.