# Why Pretraining Fails to Share Cross-Lingual Knowledge

arXiv: https://arxiv.org/abs/2609.19291 · HTML: https://mlsanity.com/paper/2609.19291/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 55 of its 68 references that are in the map.
- First posted on arXiv 16 Sep 2026.

## Builds on (most cited first, 30 of 55)

- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Similarity of Neural Network Representations Revisited](https://mlsanity.com/paper/1905.00414/) (2019)
- [The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale](https://mlsanity.com/paper/2406.17557/) (2024)
- [XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization](https://mlsanity.com/paper/2003.11080/) (2020)
- [On the Cross-lingual Transferability of Monolingual Representations](https://mlsanity.com/paper/1910.11856/) (2019)
- [Language Models are Multilingual Chain-of-Thought Reasoners](https://mlsanity.com/paper/2210.03057/) (2022)
- [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://mlsanity.com/paper/2105.13626/) (2021)
- [Large Language Models Struggle to Learn Long-Tail Knowledge](https://mlsanity.com/paper/2211.08411/) (2022)
- [How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models](https://mlsanity.com/paper/2012.15613/) (2020)
- [The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants](https://mlsanity.com/paper/2308.16884/) (2023)
- [Physics of Language Models: Part 3.1, Knowledge Storage and Extraction](https://mlsanity.com/paper/2309.14316/) (2023)
- [InfoXLM: An Information-Theoretic Framework for Cross-Lingual Language Model Pre-Training](https://mlsanity.com/paper/2007.07834/) (2020)
- [Challenges and Strategies in Cross-Cultural NLP](https://mlsanity.com/paper/2203.10020/) (2022)
- [Language Model Tokenizers Introduce Unfairness Between Languages](https://mlsanity.com/paper/2305.15425/) (2023)
- [Cross-Lingual Ability of Multilingual BERT: An Empirical Study](https://mlsanity.com/paper/1912.07840/) (2019)
- [Emerging Cross-lingual Structure in Pretrained Language Models](https://mlsanity.com/paper/1911.01464/) (2019)
- [Not All Languages Are Created Equal in LLMs: Improving Multilingual Capability by Cross-Lingual-Thought Prompting](https://mlsanity.com/paper/2305.07004/) (2023)
- [Physics of Language Models: Part 3.3, Knowledge Capacity Scaling Laws](https://mlsanity.com/paper/2404.05405/) (2024)
- [Cross-Lingual Consistency of Factual Knowledge in Multilingual Language Models](https://mlsanity.com/paper/2310.10378/) (2023)
- [How Do Large Language Models Acquire Factual Knowledge During Pretraining?](https://mlsanity.com/paper/2406.11813/) (2024)
- [HumanEval-XL: A Multilingual Code Generation Benchmark for Cross-lingual Natural Language Generalization](https://mlsanity.com/paper/2402.16694/) (2024)
- [TorchTitan: One-stop PyTorch native solution for production ready LLM pre-training](https://mlsanity.com/paper/2410.06511/) (2024)
- [Dynamic Chunking for End-to-End Hierarchical Sequence Modeling](https://mlsanity.com/paper/2507.07955/) (2025)
- [Towards a Common Understanding of Contributing Factors for Cross-Lingual Transfer in Multilingual Language Models: A Review](https://mlsanity.com/paper/2305.16768/) (2023)
- [Lost in Multilinguality: Dissecting Cross-lingual Factual Inconsistency in Transformer Language Models](https://mlsanity.com/paper/2504.04264/) (2025)
- [Revisiting the Platonic Representation Hypothesis: An Aristotelian View](https://mlsanity.com/paper/2602.14486/) (2026)
- [Enhancing Multilingual LLM Pretraining with Model-Based Data Selection](https://mlsanity.com/paper/2502.10361/) (2025)
- [RomanSetu: Efficiently unlocking multilingual capabilities of Large Language Models via Romanization](https://mlsanity.com/paper/2401.14280/) (2024)
- [ECLeKTic: a Novel Challenge Set for Evaluation of Cross-Lingual Knowledge Transfer](https://mlsanity.com/paper/2502.21228/) (2025)

## Abstract (arXiv)

Large Language Models (LLMs) have made remarkable progress in the processing and modeling of many languages. Yet, unlike human multilinguals, they exhibit surprisingly limited cross-lingual knowledge transfer. While this limitation is well documented, its origins during multilingual training remain unclear. We pretrain 360M- and 7B-parameter LLMs and show that poor cross-lingual knowledge generalization emerges during pretraining and persists under standard interventions. To isolate its cause, we employ a controlled bilingual pretraining setting using two copies of the same language, sharing identical text and token segmentation, but mapped to disjoint token spaces. We find that disjoint tokens alone are enough to induce knowledge compartmentalization, even between identical copies of the same language, establishing disjoint token spaces as a fundamental barrier to cross-lingual knowledge generalization. Guided by this understanding, we suggest mapping languages into a shared token space by simple word-wise translation and find it substantially improves cross-lingual knowledge generalization, recovering up to 12.6% of native-language learning efficiency --- 14× the baseline.