# Pretraining Language Models Using Translationese

arXiv: https://arxiv.org/abs/2403.13638 · HTML: https://mlsanity.com/paper/2403.13638/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 37 of its 62 references that are in the map.
- First posted on arXiv 20 Mar 2024.

## Builds on (most cited first, 30 of 37)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [mT5: A massively multilingual pre-trained text-to-text transformer](https://mlsanity.com/paper/2010.11934/) (2020)
- [Get To The Point: Summarization with Pointer-Generator Networks](https://mlsanity.com/paper/1704.04368/) (2017)
- [COMET: A Neural Framework for MT Evaluation](https://mlsanity.com/paper/2009.09025/) (2020)
- [Neural Network Acceptability Judgments](https://mlsanity.com/paper/1805.12471/) (2018)
- [XNLI: Evaluating Cross-lingual Sentence Representations](https://mlsanity.com/paper/1809.05053/) (2018)
- [Abstractive Text Summarization Using Sequence-to-Sequence RNNs and Beyond](https://mlsanity.com/paper/1602.06023/) (2016)
- [Sequence-Level Knowledge Distillation](https://mlsanity.com/paper/1606.07947/) (2016)
- [Improving Neural Machine Translation Models with Monolingual Data](https://mlsanity.com/paper/1511.06709/) (2015)
- [BARTScore: Evaluating Generated Text as Text Generation](https://mlsanity.com/paper/2106.11520/) (2021)
- [TinyStories: How Small Can Language Models Be and Still Speak Coherent English?](https://mlsanity.com/paper/2305.07759/) (2023)
- [XL-Sum: Large-Scale Multilingual Abstractive Summarization for 44 Languages](https://mlsanity.com/paper/2106.13822/) (2021)
- [Few-shot Learning with Multilingual Language Models](https://mlsanity.com/paper/2112.10668/) (2021)
- [Estimating the Carbon Footprint of BLOOM, a 176B Parameter Language Model](https://mlsanity.com/paper/2211.02001/) (2022)
- [DialogSum: A Real-Life Scenario Dialogue Summarization Dataset](https://mlsanity.com/paper/2105.06762/) (2021)
- [MADLAD-400: A Multilingual And Document-Level Large Audited Dataset](https://mlsanity.com/paper/2309.04662/) (2023)
- [IndicTrans2: Towards High-Quality and Accessible Machine Translation Models for all 22 Scheduled Indian Languages](https://mlsanity.com/paper/2305.16307/) (2023)
- [Towards a Cleaner Document-Oriented Multilingual Crawled Corpus](https://mlsanity.com/paper/2201.06642/) (2022)
- [Samanantar: The Largest Publicly Available Parallel Corpora Collection for 11 Indic Languages](https://mlsanity.com/paper/2104.05596/) (2021)
- [The Fact Extraction and VERification (FEVER) Shared Task](https://mlsanity.com/paper/1811.10971/) (2018)
- [Towards Leaving No Indic Language Behind: Building Monolingual Corpora, Benchmark and Models for Indic Languages](https://mlsanity.com/paper/2212.05409/) (2022)
- [Attaining the Unattainable? Reassessing Claims of Human Parity in Neural Machine Translation](https://mlsanity.com/paper/1808.10432/) (2018)
- [IndicNLG Benchmark: Multilingual Datasets for Diverse NLG Tasks in Indic Languages](https://mlsanity.com/paper/2203.05437/) (2022)
- [Translationese in Machine Translation Evaluation](https://mlsanity.com/paper/1906.09833/) (2019)

## Built on by (most cited first, 6 of 6)

- [Sailor2: Sailing in South-East Asia with Inclusive Multilingual LLMs](https://mlsanity.com/paper/2502.12982/) (2025)
- [Pula: Training Large Language Models for Setswana](https://mlsanity.com/paper/2408.02239/) (2024)
- [Regional Tiny Stories: Using Small Models to Compare Language Learning and Tokenizer Performance](https://mlsanity.com/paper/2504.07989/) (2025)
- [KletterMix: Climbing Toward High-Quality German Pretraining Data - The Full Report](https://mlsanity.com/paper/2606.03773/) (2026)
- [MameLoshnLM: Yiddish Language Model and Evaluation Benchmark](https://mlsanity.com/paper/2608.05850/) (2026)
- [Why Pretraining Fails to Share Cross-Lingual Knowledge](https://mlsanity.com/paper/2609.19291/) (2026)

## Abstract (arXiv)

In this paper, we explore the utility of translationese as synthetic data created using machine translation for pre-training language models (LMs) for low-resource languages (LRLs). Our simple methodology consists of translating large amounts of web-crawled monolingual documents (clean) into the LRLs, followed by filtering the translated documents using tiny LMs trained on small but clean LRL data. Taking the case of Indian languages, we pre-train LMs from scratch with 28M and 85M parameters, and then fine-tune them for 5 downstream natural language understanding (NLU) and 4 generative (NLG) tasks. We observe that pre-training on filtered synthetic data leads to relative performance drops of only 0.87% for NLU and 2.35% for NLG, compared to pre-training on clean data, and this gap further diminishes upon the inclusion of a small amount of clean data. We also study the impact of synthetic data filtering and the choice of source language for synthetic data generation. Furthermore, evaluating continually pre-trained larger models like Gemma-2B and Llama-3-8B in few-shot settings, we observe that using synthetic data is competitive with using clean data. Our findings suggest that synthetic data shows promise for bridging the pre-training gap between English and LRLs.