# Breaking Language Barriers: Cross-Lingual Continual Pre-Training at Scale

arXiv: https://arxiv.org/abs/2407.02118 · HTML: https://mlsanity.com/paper/2407.02118/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 41 references that are in the map.
- First posted on arXiv 2 Jul 2024.

## Builds on (most cited first, 29 of 29)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing](https://mlsanity.com/paper/1808.06226/) (2018)
- [How transferable are features in deep neural networks?](https://mlsanity.com/paper/1411.1792/) (2014)
- [A Comprehensive Survey on Transfer Learning](https://mlsanity.com/paper/1911.02685/) (2019)
- [XNLI: Evaluating Cross-lingual Sentence Representations](https://mlsanity.com/paper/1809.05053/) (2018)
- [A Survey on Deep Transfer Learning](https://mlsanity.com/paper/1808.01974/) (2018)
- [Orthogonal Gradient Descent for Continual Learning](https://mlsanity.com/paper/1910.07104/) (2019)
- [XCOPA: A Multilingual Dataset for Causal Commonsense Reasoning](https://mlsanity.com/paper/2005.00333/) (2020)
- [Multilingual Translation with Extensible Multilingual Pretraining and Finetuning](https://mlsanity.com/paper/2008.00401/) (2020)
- [Few-shot Learning with Multilingual Language Models](https://mlsanity.com/paper/2112.10668/) (2021)
- [Scaling Laws for Transfer](https://mlsanity.com/paper/2102.01293/) (2021)
- [WinoGrande: An Adversarial Winograd Schema Challenge at Scale](https://mlsanity.com/paper/1907.10641/) (2019)
- [When Scaling Meets LLM Finetuning: The Effect of Data, Model and Finetuning Method](https://mlsanity.com/paper/2402.17193/) (2024)
- [Okapi: Instruction-tuned Large Language Models in Multiple Languages with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2307.16039/) (2023)
- [Emerging Cross-lingual Structure in Pretrained Language Models](https://mlsanity.com/paper/1911.01464/) (2019)
- [Scaling Laws for Fine-Grained Mixture of Experts](https://mlsanity.com/paper/2402.07871/) (2024)
- [Fine-tuned Language Models are Continual Learners](https://mlsanity.com/paper/2205.12393/) (2022)
- [Scaling Laws and Interpretability of Learning from Repeated Data](https://mlsanity.com/paper/2205.10487/) (2022)
- [Continual Pre-Training of Large Language Models: How to (re)warm your model?](https://mlsanity.com/paper/2308.04014/) (2023)
- [Scaling Laws vs Model Architectures: How does Inductive Bias Influence Scaling?](https://mlsanity.com/paper/2207.10551/) (2022)
- [Continual Learning with Foundation Models: An Empirical Study of Latent Replay](https://mlsanity.com/paper/2205.00329/) (2022)
- [Scaling Laws for Sparsely-Connected Foundation Models](https://mlsanity.com/paper/2309.08520/) (2023)
- [Zero-shot cross-lingual transfer language selection using linguistic similarity](https://mlsanity.com/paper/2301.13720/) (2023)
- [Overcoming Catastrophic Forgetting in Massively Multilingual Continual Learning](https://mlsanity.com/paper/2305.16252/) (2023)

## Built on by (most cited first, 15 of 15)

- [SEA-LION: Southeast Asian Languages in One Network](https://mlsanity.com/paper/2504.05747/) (2025)
- [Rethinking Multilingual Continual Pretraining: Data Mixing for Adapting LLMs Across Languages and Resources](https://mlsanity.com/paper/2504.04152/) (2025)
- [Bring Your Own Knowledge: A Survey of Methods for LLM Knowledge Expansion](https://mlsanity.com/paper/2502.12598/) (2025)
- [Bhaasha, Bhasa, Zaban: A Survey for Low-Resourced Languages in South Asia -- Current Stage and Challenges](https://mlsanity.com/paper/2509.11570/) (2025)
- [Beyond Static Models: An Evolving Framework for Continual Learning in Large Language Models across Training Stages](https://mlsanity.com/paper/2603.12658/) (2026)
- [Enhancing LLM Language Adaption through Cross-lingual In-Context Pre-training](https://mlsanity.com/paper/2504.20484/) (2025)
- [Full-Parameter Continual Pretraining of Gemma2: Insights into Fluency and Domain Knowledge](https://mlsanity.com/paper/2505.05946/) (2025)
- [Mitigating Catastrophic Forgetting in Target Language Adaptation of LLMs via Source-Shielded Updates](https://mlsanity.com/paper/2512.04844/) (2025)
- [OpenSeal: Good, Fast, and Cheap Construction of an Open-Source Southeast Asian LLM via Parallel Data](https://mlsanity.com/paper/2602.02266/) (2026)
- [The Rise and Down of Babel Tower: Investigating the Evolution Process of Multilingual Code Large Language Model](https://mlsanity.com/paper/2412.07298/) (2024)
- [CultureMERT: Continual Pre-Training for Cross-Cultural Music Representation Learning](https://mlsanity.com/paper/2506.17818/) (2025)
- [FTibSuite: A Comprehensive Resource Suite for Tibetan Vision-Language Modeling](https://mlsanity.com/paper/2605.26601/) (2026)
- [Scaling Laws for Conditional Emergence of Multilingual Image Captioning via Generalization from Translation](https://mlsanity.com/paper/2503.09443/) (2025)
- [Repetition over Diversity: High-Signal Data Filtering for Sample-Efficient German Language Modeling](https://mlsanity.com/paper/2604.28075/) (2026)
- [Why Low-Resource NLP Needs More Than Cross-Lingual Transfer: Lessons Learned from Luxembourgish](https://mlsanity.com/paper/2605.10714/) (2026)

## Abstract (arXiv)

In recent years, Large Language Models (LLMs) have made significant strides towards Artificial General Intelligence. However, training these models from scratch requires substantial computational resources and vast amounts of text data. In this paper, we explore an alternative approach to constructing an LLM for a new language by continually pretraining (CPT) from existing pretrained LLMs, instead of using randomly initialized parameters. Based on parallel experiments on 40 model sizes ranging from 40M to 5B parameters, we find that 1) CPT converges faster and saves significant resources in a scalable manner; 2) CPT adheres to an extended scaling law derived from Hoffmann et al. (2022) with a joint data-parameter scaling term; 3) The compute-optimal data-parameter allocation for CPT markedly differs based on our estimated scaling factors; 4) The effectiveness of transfer at scale is influenced by training duration and linguistic properties, while robust to data replaying, a method that effectively mitigates catastrophic forgetting in CPT. We hope our findings provide deeper insights into the transferability of LLMs at scale for the research community.