Multilingual Neural Machine Translation

Multilingual neural machine translation (MNMT) aims to build single models capable of translating between numerous language pairs, improving efficiency and resource allocation compared to training separate bilingual models. Current research focuses on optimizing model architectures (like Mixture-of-Experts) and training strategies to address challenges such as parameter inefficiency, negative interactions between languages during fine-tuning, and the "off-target" problem (incorrect language output). These advancements are significant because they enable more efficient and effective translation for low-resource languages and improve the overall quality and robustness of machine translation systems.

Papers

June 22, 2023

Multilingual Neural Machine Translation System for Indic to Indic Languages
Sudhansu Bala Das, Divyajyoti Panda, Tapas Kumar Mishra, Bidyut Kr. Patra, Asif Ekbal
Indian Language Multilingual Neural Machine Translation Tamil Language Indian English

June 12, 2023

Learning Multilingual Sentence Representations with Cross-lingual Consistency Regularization
Pengzhi Gao, Liwen Zhang, Zhongjun He, Hua Wu, Haifeng Wang
Multilingual Neural Machine Translation Task Consistency Multilingual Sentence Multilingual Sentence Encoders

May 31, 2023

TPDM: Selectively Removing Positional Information for Zero-shot Translation via Token-Level Position Disentangle Module
Xingran Chen, Ge Zhang, Jie Fu
Token Level Multilingual Neural Machine Translation Zero Shot Translation Positional Information

May 26, 2023

Gender Lost In Translation: How Bridging The Gap Between Languages Affects Gender Bias in Zero-Shot Multilingual Translation
Lena Cabrera, Jan Niehues
Zero Shot Gender Bias View Translation Gender Information Multilingual Neural Machine Translation Zero Shot Translation Language Agnostic Multilingual Translation

May 25, 2023

Towards Higher Pareto Frontier in Multilingual Machine Translation
Yichong Huang, Xiaocheng Feng, Xinwei Geng, Baohang Li, Bing Qin
Knowledge Distillation Multilingual Machine Translation Multilingual Neural Machine Translation Multilingual Corpus Pareto Frontier Pareto Set

May 23, 2023

Beyond Shared Vocabulary: Increasing Representational Word Similarities across Languages for Multilingual Machine Translation
Di Wu, Christof Monz
Jina Embeddings Word Embeddings Unknown Language Multilingual Machine Translation Multilingual Neural Machine Translation Representational Similarity Word Level Equivalence Class

May 22, 2023

Mitigating Data Imbalance and Representation Degeneration in Multilingual Machine Translation
Wen Lai, Alexandra Chronopoulou, Alexander Fraser
Parallel Corpus Language Pair Multilingual Machine Translation Multilingual Neural Machine Translation Data Imbalance Monolingual Data Representation Degeneration

May 21, 2023

Communication Efficient Federated Learning for Multilingual Neural Machine Translation with Adapter
Yi Liu, Xiaohan Bi, Lei Li, Sishuo Chen, Wenkai Yang, Xu Sun
Pre Trained Language Model Multilingual Neural Machine Translation Communication Efficient Federated Learning Long Short Range Adapter Federated Large Language Model

May 19, 2023

Viewing Knowledge Transfer in Multilingual Machine Translation Through a Representational Lens
David Stap, Vlad Niculae, Christof Monz
Knowledge Transfer Translation Quality Multilingual Machine Translation Multilingual Neural Machine Translation Transferable Representation Representational Similarity

May 18, 2023

On the Off-Target Problem of Zero-Shot Multilingual Neural Machine Translation
Liang Chen, Shuming Ma, Dongdong Zhang, Furu Wei, Baobao Chang
Multilingual Neural Machine Translation Zero Shot Translation Translation Benchmark Zero Shot Multilingual

May 17, 2023

Variable-length Neural Interlingua Representations for Zero-shot Neural Machine Translation
Zhuoyuan Mao, Haiyue Song, Raj Dabre, Chenhui Chu, Sadao Kurohashi
Multilingual Neural Machine Translation Zero Shot Translation

May 12, 2023

Improving Zero-shot Multilingual Neural Machine Translation by Leveraging Cross-lingual Consistency Regularization
Pengzhi Gao, Liwen Zhang, Zhongjun He, Hua Wu, Haifeng Wang
Multilingual Neural Machine Translation Zero Shot Translation Task Consistency Zero Shot Multilingual

May 11, 2023

Chain-of-Dictionary Prompting Elicits Translation in Large Language Models
Hongyuan Lu, Haoyang Huang, Dongdong Zhang, Haoran Yang, Wai Lam, Furu Wei
Large Language Model Low Resource Language Cross Lingual Multilingual Neural Machine Translation Bilingual Dictionary Translation Capability

April 19, 2023

An Empirical Study of Leveraging Knowledge Distillation for Compressing Multilingual Neural Machine Translation Models
Varun Gumma, Raj Dabre, Pratyush Kumar
Knowledge Distillation Empirical Study Multilingual Neural Machine Translation Deeper Model Distilled Model

April 16, 2023

Neural Machine Translation For Low Resource Languages
Vakul Goyle, Parvathy Krishnaswamy, Kannan Girija Ravikumar, Utsa Chattopadhyay, Kartikay Goyle
Transfer Learning Neural Machine Translation Low Resource Language Neural Machine Translation Model Multilingual Neural Machine Translation

April 6, 2023

On the Pareto Front of Multilingual Neural Machine Translation
Liang Chen, Shuming Ma, Dongdong Zhang, Furu Wei, Baobao Chang
Multilingual Model Training Corpus Multilingual Neural Machine Translation Pareto Front Multi Task Optimization

March 31, 2023

February 19, 2023

Scaling Laws for Multilingual Neural Machine Translation
Patrick Fernandes, Behrooz Ghorbani, Xavier Garcia, Markus Freitag, Orhan Firat
Multilingual Model Scaling Law Multilingual Neural Machine Translation Language Similarity

December 20, 2022

Lego-MT: Learning Detachable Models for Massively Multilingual Machine Translation
Fei Yuan, Yinquan Lu, WenHao Zhu, Lingpeng Kong, Lei Li, Yu Qiao, Jingjing Xu
Multilingual Machine Translation Multilingual Neural Machine Translation Translation Benchmark Effective Model

Multilingual Neural Machine Translation

Papers

Multilingual Neural Machine Translation System for Indic to Indic Languages

Learning Multilingual Sentence Representations with Cross-lingual Consistency Regularization

TPDM: Selectively Removing Positional Information for Zero-shot Translation via Token-Level Position Disentangle Module

Gender Lost In Translation: How Bridging The Gap Between Languages Affects Gender Bias in Zero-Shot Multilingual Translation

Towards Higher Pareto Frontier in Multilingual Machine Translation

Beyond Shared Vocabulary: Increasing Representational Word Similarities across Languages for Multilingual Machine Translation

Mitigating Data Imbalance and Representation Degeneration in Multilingual Machine Translation

Communication Efficient Federated Learning for Multilingual Neural Machine Translation with Adapter

Viewing Knowledge Transfer in Multilingual Machine Translation Through a Representational Lens

On the Off-Target Problem of Zero-Shot Multilingual Neural Machine Translation

Variable-length Neural Interlingua Representations for Zero-shot Neural Machine Translation

Improving Zero-shot Multilingual Neural Machine Translation by Leveraging Cross-lingual Consistency Regularization

Chain-of-Dictionary Prompting Elicits Translation in Large Language Models

An Empirical Study of Leveraging Knowledge Distillation for Compressing Multilingual Neural Machine Translation Models

Neural Machine Translation For Low Resource Languages

On the Pareto Front of Multilingual Neural Machine Translation

Exploiting Multilingualism in Low-resource Neural Machine Translation via Adversarial Learning

$\varepsilon$ K\'U <MASK>: Integrating Yor\`ub\'a cultural greetings into machine translation

Scaling Laws for Multilingual Neural Machine Translation

Lego-MT: Learning Detachable Models for Massively Multilingual Machine Translation