PaperScope
LIVE · 2026-09-11 05:40 UTC

Structural priors for data-efficient language learning

Yana Veitsman, Jonas Mayer Martins, Jonathan Lautenschlager, Lisa Beinborn

Latestcs.CLcs.LGcs.AIcs.CV
arXiv ID
2609.11505 v1
Category
Submitted
2026-09-10

Abstract

Efficient language learning requires methods to reduce the reliance on large data and computational resources. We investigate structural transfer: First training models on non-language data to induce useful priors for natural language. This approach is a form of weight initialization for multilingual language modeling. We evaluate transfer via next-token-prediction loss, weight shifts in the model, and downstream linguistic benchmarks. Several symbolic data types - notably music, probabilistic grammars, and cellular automata - yield lower language-modeling loss than random initialization. These gains coincide with smaller weight shifts during subsequent language training, suggesting that structural transfer positions models in a more favorable region of the parameter space. However, a lower loss does not translate consistently into better downstream linguistic performance, and transfer from non-language data is less efficient than additional language data. We conclude that non-language data can serve as a partial substitute for language data for the training objective of next-token prediction but does not reliably support broader linguistic generalization.

Comment: EMNLP 2026, BabyLM Challenge; 18 pages, 11 figures

arXiv abs page · PDF · same-day batch