m4Adapter: Multilingual Multi-Domain Adaptation for Machine Translation with a Meta-Adapter
Not on the map. Why?
Abstract (arXiv)
Multilingual neural machine translation models (MNMT) yield state-of-the-art performance when evaluated on data from a domain and language pair seen at training time. However, when a MNMT model is used to translate under domain shift or to a new language pair, performance drops dramatically. We consider a very challenging scenario: adapting the MNMT model both to a new domain and to a new language pair at the same time. In this paper, we propose m4Adapter (Multilingual Multi-Domain Adaptation for Machine Translation with a Meta-Adapter), which combines domain and language knowledge using meta-learning with adapters. We present results showing that our approach is a parameter-efficient solution which effectively adapts a model to both a new language pair and a new domain, while outperforming other adapter methods. An ablation study also shows that our approach more effectively transfers domain knowledge across different languages and language information across different domains.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 36 of its 51 references that are in the map. First posted on arXiv 21 Oct 2022.
Builds on
36 of 51 references.
References · 36
- 2017Attention Is All You Needcited by 59,932
- 2016Layer Normalizationcited by 5,579
- 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
- 2019Parameter-Efficient Transfer Learning for NLPcited by 4,195
- 2018SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processingcited by 1,973
- 2018A Call for Clarity in Reporting BLEU Scorescited by 1,790
- 2018On First-Order Meta-Learning Algorithmscited by 1,003
- 2015Improving Neural Machine Translation Models with Monolingual Datacited by 784
- 2020Beyond English-Centric Multilingual Machine Translationcited by 559
- 2016Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translationcited by 467
- 2020AdapterHub: A Framework for Adapting Transformerscited by 441
- 2020Improving Massively Multilingual Neural Machine Translation and Zero-Shot Translationcited by 277
24 more
- 2019Massively Multilingual Neural Machine Translationcited by 246
- 2019Simple, Scalable Adaptation for Neural Machine Translationcited by 239
- 2020AdapterDrop: On the Efficiency of Adapters in Transformerscited by 223
- 2021Meta-learning via Language Model In-context Tuningcited by 163
- 2020Unsupervised Domain Clusters in Pretrained Language Modelscited by 160
- 2020Tangled up in BLEU: Reevaluating the Evaluation of Automatic Machine Translation Evaluation Metricscited by 155
- 2018Meta-Learning for Low-Resource Neural Machine Translationcited by 105
- 2018Rapid Adaptation of Neural Machine Translation to New Languagescited by 82
- 2018A Survey of Domain Adaptation for Neural Machine Translationcited by 64
- 2019Improved Zero-shot Neural Machine Translation via Ignoring Spurious Correlationscited by 61
- 2019Investigating Meta-Learning Algorithms for Low-Resource Natural Language Understanding Taskscited by 48
- 2016Domain Control for Neural Machine Translationcited by 36
- 2017Dynamic Data Selection for Neural Machine Translationcited by 33
- 2016Fast Domain Adaptation for Neural Machine Translationcited by 28
- 2020Recipes for Adapting Pre-trained Monolingual and Multilingual Models to Machine Translationcited by 28
- 2021Meta-Curriculum Learning for Domain Adaptation in Neural Machine Translationcited by 28
- 2022Meta Learning for Natural Language Processing: A Surveycited by 26
- 2021Multilingual Domain Adaptation for NMT: Decoupling Language and Domain Information with Adapterscited by 21
- 2020Meta-Learning for Few-Shot NMT Adaptationcited by 20
- 2021Meta-Learning for Effective Multi-task and Multilingual Modellingcited by 13
- 2021Improving Both Domain Robustness and Domain Adaptability in Machine Translationcited by 11
- 2019Multilingual Multi-Domain Adaptation Approaches for Neural Machine Translationcited by 8
- 2022DaLC: Domain Adaptation Learning Curve Prediction for Neural Machine Translationcited by 2
- 2017Building a Neural Machine Translation System Using Only Synthetic Parallel Datacited by 1
Built on by
Citing papers per year: 2022: 1; 2023: 1.
2 papers.