mlsanityData to 14 Oct 2026

A Factorial Study of Synthetic Data Generation for Low-Resource Machine Translation using Grammar Books

arXiv 2607.22376 ↗First posted 24 Jul 2026cs.CLcited by 0

On the map: Synthetic data and self-improvement for LLMs → LLM synthetic data generation

Abstract (arXiv)

Most endangered languages lack the parallel data required for machine translation, despite the existence of descriptive grammar books. We introduce a pipeline that uses large language models to extract grammatical rules, example sentences, and lexicons from grammar books and generate synthetic parallel corpora for fine-tuning-rather than feeding grammar content into prompts at inference time, as in prior work. Validated on three typologically diverse low-resource languages-Kalamang (Papuan), Tuatschin (Romance), and Mandan (Siouan)-we show that fine-tuning on synthetic data improves over seed-data baselines in 75% of configurations for Kalamang and 59% for Tuatschin, with best-case ChrF++ gains of +8.8, +5.3, and +3.3 respectively. Through a systematic factorial study across 96 configurations varying target part-of-speech, retrieval granularity, and sample volume, we identify which factor combinations drive gains and where they break down. Our results demonstrate that static linguistic documentation can be repurposed for machine translation fine-tuning, offering a practical path towards translation tools for severely under-resourced languages.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in LLM synthetic data generation, one of 13 directions in Synthetic data and self-improvement for LLMs. Builds on 15 of its 38 references that are in the map. First posted on arXiv 24 Jul 2026.

Builds on

15 of 38 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025Scaling Low-Resource MT via Synthetic Data Generation with LLMs20 May

Other references · 14

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2022No Language Left Behind: Scaling Human-Centered Machine Translationcited by 1,312
  3. 2015Improving Neural Machine Translation Models with Monolingual Datacited by 788
  4. 2017Data Augmentation for Low-Resource Neural Machine Translationcited by 139
  5. 2021Survey of Low-Resource Machine Translationcited by 92
  6. 2023A Benchmark for Learning to Translate a New Language from One Grammar Bookcited by 75
  7. 2024Teaching Large Language Models an Unseen Language on the Flycited by 30
  8. 2024Can LLMs Really Learn to Translate a Low-Resource Language from One Grammar Book?cited by 23
  9. 2025Compositional Translation: A Novel LLM-based Approach for Low-resource Machine Translationcited by 14
  10. 2024Back to School: Translation Using Grammar Bookscited by 11
  11. 2024Can LLMs Help Create Grammar?: Automating Grammar Creation for Endangered Languages with In-Context Learningcited by 8
  12. 2024Rule-Based, Neural and LLM Back-Translation: Comparative Insights from a Variant of Ladincited by 5
2 more
  1. 2025Read it in Two Steps: Translating Extremely Low-Resource Languages with Code-Augmented Grammar Bookscited by 4
  2. 2024A Morphologically-Aware Dictionary-based Data Augmentation Technique for Machine Translation of Under-Represented Languagescited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025BhashaKritika: Building Synthetic Pretraining Data at Scale for Indic Languages
  2. 2025Scaling Laws of Synthetic Data for Language Models
  3. 2025Contrastive Decoding for Synthetic Data Generation in Low-Resource Language Modeling
  4. 2026Reasoning over Grammar: Can Synthetic Linguistic Reasoning Traces Enhance Low-Resource Machine Translation?
  5. 2026Better as Generators Than Classifiers: Leveraging LLMs and Synthetic Data for Low-Resource Multilingual Classification
  6. 2023Synthetic Data Generation in Low-Resource Settings via Fine-Tuning of Large Language Models
  7. 2025Bridging the Linguistic Divide: A Survey on Leveraging Large Language Models for Machine Translation
  8. 2025Demystifying Synthetic Data in LLM Pre-training: A Systematic Study of Scaling Laws, Benefits, and Pitfalls