# Multilingual Translation with Extensible Multilingual Pretraining and Finetuning

arXiv: https://arxiv.org/abs/2008.00401 · HTML: https://mlsanity.com/paper/2008.00401/

## Key facts

- Cited by 287 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Aug 2020.

## Built on by (most cited first, 30 of 287)

- [The FLORES-101 Evaluation Benchmark for Low-Resource and Multilingual Machine Translation](https://mlsanity.com/paper/2106.03193/) (2021)
- [Language Model Tokenizers Introduce Unfairness Between Languages](https://mlsanity.com/paper/2305.15425/) (2023)
- [SIB-200: A Simple, Inclusive, and Big Evaluation Dataset for Topic Classification in 200+ Languages and Dialects](https://mlsanity.com/paper/2309.07445/) (2023)
- [Adapting Pre-trained Language Models to African Languages via Multilingual Adaptive Fine-Tuning](https://mlsanity.com/paper/2204.06487/) (2022)
- [Samanantar: The Largest Publicly Available Parallel Corpora Collection for 11 Indic Languages](https://mlsanity.com/paper/2104.05596/) (2021)
- [Accelerating Transformer Inference for Translation via Parallel Decoding](https://mlsanity.com/paper/2305.10427/) (2023)
- [A Few Thousand Translations Go a Long Way! Leveraging Pre-trained Models for African News Translation](https://mlsanity.com/paper/2205.02022/) (2022)
- [Survey of Low-Resource Machine Translation](https://mlsanity.com/paper/2109.00486/) (2021)
- [Bactrian-X: Multilingual Replicable Instruction-Following Models with Low-Rank Adaptation](https://mlsanity.com/paper/2305.15011/) (2023)
- [Multilingual Large Language Models Are Not (Yet) Code-Switchers](https://mlsanity.com/paper/2305.14235/) (2023)
- [Collecting a Large-Scale Gender Bias Dataset for Coreference Resolution and Machine Translation](https://mlsanity.com/paper/2109.03858/) (2021)
- [UnitY: Two-pass Direct Speech-to-speech Translation with Discrete Units](https://mlsanity.com/paper/2212.08055/) (2022)
- [AVATAR: A Parallel Corpus for Java-Python Program Translation](https://mlsanity.com/paper/2108.11590/) (2021)
- [IndoNLG: Benchmark and Resources for Evaluating Indonesian Natural Language Generation](https://mlsanity.com/paper/2104.08200/) (2021)
- [Facebook AI WMT21 News Translation Task Submission](https://mlsanity.com/paper/2108.03265/) (2021)
- [Embarrassingly Easy Document-Level MT Metrics: How to Convert Any Pretrained Metric Into a Document-Level Metric](https://mlsanity.com/paper/2209.13654/) (2022)
- [Lightweight Adapter Tuning for Multilingual Speech Translation](https://mlsanity.com/paper/2106.01463/) (2021)
- [DeltaLM: Encoder-Decoder Pre-training for Language Generation and Translation by Augmenting Pretrained Multilingual Encoders](https://mlsanity.com/paper/2106.13736/) (2021)
- [Towards Building ASR Systems for the Next Billion Users](https://mlsanity.com/paper/2111.03945/) (2021)
- [EHRSQL: A Practical Text-to-SQL Benchmark for Electronic Health Records](https://mlsanity.com/paper/2301.07695/) (2023)
- [Democratizing LLMs for Low-Resource Languages by Leveraging their English Dominant Abilities with Linguistically-Diverse Prompts](https://mlsanity.com/paper/2306.11372/) (2023)
- [Multilingual training for Software Engineering](https://mlsanity.com/paper/2112.02043/) (2021)
- [Improving Language Plasticity via Pretraining with Active Forgetting](https://mlsanity.com/paper/2307.01163/) (2023)
- [BanglaNLG and BanglaT5: Benchmarks and Resources for Evaluating Low-Resource Natural Language Generation in Bangla](https://mlsanity.com/paper/2205.11081/) (2022)
- [Multilingual Generative Language Models for Zero-Shot Cross-Lingual Event Argument Extraction](https://mlsanity.com/paper/2203.08308/) (2022)
- [IndicBART: A Pre-trained Model for Indic Natural Language Generation](https://mlsanity.com/paper/2109.02903/) (2021)
- [The Effect of Domain and Diacritics in Yor`ub'a-English Neural Machine Translation](https://mlsanity.com/paper/2103.08647/) (2021)
- [A Survey on Cross-Lingual Summarization](https://mlsanity.com/paper/2203.12515/) (2022)
- [Models and Datasets for Cross-Lingual Summarisation](https://mlsanity.com/paper/2202.09583/) (2022)
- [Memory-efficient NLLB-200: Language-specific Expert Pruning of a Massively Multilingual Machine Translation Model](https://mlsanity.com/paper/2212.09811/) (2022)

## Abstract (arXiv)

Recent work demonstrates the potential of multilingual pretraining of creating one model that can be used for various tasks in different languages. Previous work in multilingual pretraining has demonstrated that machine translation systems can be created by finetuning on bitext. In this work, we show that multilingual translation models can be created through multilingual finetuning. Instead of finetuning on one direction, a pretrained model is finetuned on many directions at the same time. Compared to multilingual models trained from scratch, starting from pretrained models incorporates the benefits of large quantities of unlabeled monolingual data, which is particularly important for low resource languages where bitext is not available. We demonstrate that pretrained models can be extended to incorporate additional languages without loss of performance. We double the number of languages in mBART to support multilingual machine translation models of 50 languages. Finally, we create the ML50 benchmark, covering low, mid, and high resource languages, to facilitate reproducible research by standardizing training and evaluation data. On ML50, we demonstrate that multilingual finetuning improves on average 1 BLEU over the strongest baselines (being either multilingual from scratch or bilingual finetuning) while improving 9.3 BLEU on average over bilingual baselines from scratch.