arXiv AI

Why Pretraining Fails to Share Cross-Lingual Knowledge

The paper investigates why large language models (LLMs) fail to transfer knowledge across languages. By pretraining 360M- and 7B-parameter models, the authors show that poor cross‑lingual generalization arises during pretraining and persists despite standard fixes. Using a controlled bilingual setup with identical text but disjoint token spaces, they demonstrate that token disjointness alone causes knowledge compartmentalization, and that mapping languages into a shared token space via word‑wise translation markedly improves cross‑lingual performance, recovering up to 12.6% of native‑language learning efficiency.

arXiv Computation and Language
Oct 12

Copy First, Translate Later: Interpreting Translation Dynamics in Multilingual Pretraining

Large language models show strong cross‑lingual abilities, but previous studies have examined them only at isolated training points. This work pretrains a 1.7B multilingual model on nine languages, recording checkpoints at fine granularity to track translation development. Using word‑level translation and a new dataset, the authors find that early learning is dominated by token‑level copying and surface similarities, while later stages involve more general translation mechanisms and refined copying.

By Felicia K\"orner, Maria Matveev, Florian Eichin, Gitta Kutyniok, Barbara Plank, Michael A. Hedderich
arXiv Computation and Language
Oct 22

How to Use Parallel Data for Multilingual Pretraining

The paper explores how to leverage parallel corpora for multilingual pretraining by concatenating each document with its translation in the same training sequence, a method called Parallel Pretraining (PPT). Compared to standard pretraining (SPT), PPT alone does not consistently improve language understanding, but it builds a stronger cross‑lingual foundation. A two‑stage approach—first PPT to establish cross‑lingual knowledge, then SPT to fine‑tune for standard language modeling—consistently outperforms standard pretraining in both high‑resource and low‑resource settings.

By Zhihong Huang, Yao Lu, Derek F. Wong, Hamid Alinejad-Rokny, Min Yang
arXiv Computation and Language
Aug 28

Double Trouble: Bilingual Pretraining Leaves Language-Conditioned Effects in Shared-Language Representations

The study compares an English-only and a bilingual decoder-only model, each 310 M parameters, trained on eight diverse languages while controlling for English exposure, compute, and document overlap. After aligning on shared English vocabulary, the authors find that token embeddings appear similar, but the deeper hidden states used for prediction diverge across models. This hidden‑state mismatch grows through middle transformer layers and persists despite controls, indicating that contextual processing differs between the models. "whyItMatters":"The findings show that embedding alignment can conceal significant internal representation differences, which is crucial for any downstream work that assumes aligned multilingual models are interchangeable."

By Anjishnu Mukherjee, Ziwei Zhu, Antonios Anastasopoulos
arXiv Computation and Language
Aug 27

One Form to Transfer Them All: Pretraining Multilingual Language Models Beyond Native Orthography

The study evaluates how different input representations—orthographic text, IPA transcription, and romanization—affect cross‑lingual transfer in autoregressive multilingual language models. Across three model sizes and eight languages grouped into typologically motivated pairs, romanized pretraining consistently outperforms native orthography and IPA, especially as model scale increases. Fine‑tuning a text‑pretrained model on romanized data can harm performance on languages already covered by the base model, suggesting romanization should be integrated at pretraining rather than applied later.

By Muge Zhang, Aaron Jencks, Krishna Badikela, Yulia Tsvetkov, Sachin Kumar
arXiv AI
Aug 28

Beyond the Rosetta Stone: Unification Forces in Generalization Dynamics

The paper investigates why large language models sometimes hallucinate when asked about facts in a language different from the one in which the facts were learned. By training small Transformer models on synthetic multilingual datasets, the authors show that the degree of correlation between facts and their learning language (informativeness) and the ease of language identification (extractability) determine whether models develop unified or separate representations across languages. Unified representations enable cross‑lingual fact transfer, while separate representations do not. The study proposes a unifying perspective on cross‑lingual transfer and suggests training methods to promote representational unification.

By Carter Blum, Katja Filippova, Ann Yuan, Asma Ghandeharioun, Julian Zimmert, Fred Zhang, Jessica Hoffmann, Tal Linzen, Martin Wattenberg, Lucas Dixon, Mor Geva
arXiv AI
Sep 4

One Model to Translate Them All? A Journey to Mount Doom for Multilingual Model Merging

The paper investigates weight‑space merging of independently fine‑tuned multilingual machine translation models. Experiments show that merging is more successful when models share a target language, yet it still cannot match the peak performance of language‑specific checkpoints. When target languages differ, performance drops sharply, and analysis reveals that overlapping neuron activation and incompatible upper‑layer geometries cause these failures.

By Baban Gain, Trilok Nath Singh, Asif Ekbal
arXiv Computation and Language
Sep 10

MultiSynt/MT: Trillion-Token Multi-Parallel Pre-Training Data Translated Across 36 Languages

arXiv:2607.00890v2 Announce Type: replace Abstract: Open web-scale pre-training corpora remain concentrated in English, limiting multilingual LLM development. We introduce MultiSynt/MT, an open synth...

By Maximilian Idahl, J\"org Tiedemann, Sampo Pyysalo, David Salinas, Tomasz Galica, Shenbin Qian, Tudor Nicolae Mateiu, Zihao Li, Anna Lokrantz, Fedor Vitiugin, Andr\'e F. T. Martins, Jenna Kanerva, Filip Ginter, Matthias Lindemann, Tim Isbister, Birger Moell, Jonas Lindh, Jan Haji\v{c}, Jenia Jitsev, Andrey Kutuzov, Stephan Oepen, Gema Ram\'irez-S\'anchez
arXiv Computation and Language
Sep 28

Feeding BabyLMs Macaroni: Code-Switching Curricula Cause Cross-Lingual Convergence

The paper investigates whether training small decoder-only transformers on code‑switched text can induce cross‑lingual alignment. Using two 100‑million‑word multilingual corpora—one a mix of English, Dutch, and Chinese BabyBabelLM data, and another generated by inserting word‑ and sentence‑level code‑switching via an LLM—the authors find that code‑switched training aligns representations of parallel text, especially across different scripts, and that this alignment persists when later training on monolingual documents. A curriculum that progresses from word‑level code‑switching to sentence‑level code‑switching and finally to monolingual data yields models that outperform baselines on the BabyLM evaluation suite, demonstrating that code‑switching curriculum learning is an effective data augmentation strategy for multilingual pretraining.

By Dries Rooryck, Alex Cai, Yonatan Belinkov, David Alvarez-Melis, Kiant\'e Brantley