Do Construction Distributions Shape Formal Language Learning In German BabyLMs?
This work addresses how linguistic stimuli influence language learning, offering insights for developmental linguistics and AI, but it is incremental as it applies existing methods to new data.
The study investigated how different distributions of constructions in German child-directed speech affect language learning in small language models, finding that learning trajectories are robust with little impact on final accuracy, though syntax learning improves with complex utterances and word-level learning with fragmentary ones.
We analyze the influence of utterance-level construction distributions in German child-directed/child-available speech on the resulting word-level, syntactic and semantic competence (and their underlying learning trajectories) in small LMs, which we train on a novel collection of developmentally plausible language data for German. We find that trajectories are surprisingly robust for markedly different distributions of constructions in the training data, which have little effect on final accuracies and almost no effect on global learning trajectories. While syntax learning benefits from more complex utterances, word-level learning culminates in better scores with more fragmentary utterances. We argue that LMs trained on developmentally plausible data can contribute to debates on how conducive different kinds of linguistic stimuli are to language learning.