Metraipolis › archive › dc_9cc0afad6dec

Feeding BabyLMs Macaroni: Code-Switching Curricula Cause Cross-Lingual Convergence — arXiv:2609.30535

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.30535
License (as archived)
CC0-1.0
Archived
2026-10-08T02:03:55.910Z
SHA-256 of the archived text
bf92c71af22c2726bbacfc65b92f68f8aa8545c5c118701fec39e045f857ffa0
Stable URL of this page
https://metraipolis.com/documento/dc_9cc0afad6dec

How to cite

Feeding BabyLMs Macaroni: Code-Switching Curricula Cause Cross-Lingual Convergence — arXiv:2609.30535 — arXiv (en), https://arxiv.org/abs/2609.30535 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9cc0afad6dec (Metraipolis archive dc_9cc0afad6dec, sha256 bf92c71af22c2726bbacfc65b92f68f8aa8545c5c118701fec39e045f857ffa0)
BibTeX
@misc{dc9cc0afad6dec,
  title = {Feeding BabyLMs Macaroni: Code-Switching Curricula Cause Cross-Lingual Convergence — arXiv:2609.30535},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.30535},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9cc0afad6dec; sha256 bf92c71af22c2726bbacfc65b92f68f8aa8545c5c118701fec39e045f857ffa0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.30535. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Children in multilingual communities often code-switch, using multiple languages in a single utterance. Can we induce cross-lingual alignment in language models by training on code-switched text? We pretrain small decoder-only transformers on two 100M-word multilingual corpora: a base corpus formed by mixing the English, Dutch, and Chinese BabyBabelLM datasets, and a corpus generated from it by inserting word- and sentence-level code-switching using an LLM. We find that training on code-switched data aligns the representations of parallel text, particularly across different scripts, and that this alignment persists through training on monolingual documents. Under a learning curriculum that progresses from word-level code-switching, to sentence-level code-switching, to monolingual documents, models trained on code-switched data outperform baselines trained without it on the BabyLM evaluation suite. Our work characterizes code-switching curriculum learning as an effective data augmentation method for multilingual pretraining. We release our code, data, and models at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.30535 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T01:38:49.428Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).