English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

When acquiring syntax, children consistently choose hierarchical rules overcompeting non-hierarchical possibilities. Is this preference due to a learningbias for hierarchical structure, or due to more general biases that interactwith hierarchical cues in children's linguistic input? We explore thesepossibilities by training LSTMs and Transformers - two types of neural networkswithout a hierarchical bias - on data similar in quantity and content tochildren's linguistic input: text from the CHILDES corpus. We then evaluatewhat these models have learned about English yes/no questions, a phenomenon forwhich hierarchical structure is crucial. We find that, though they perform wellat capturing the surface statistics of child-directed speech (as measured byperplexity), both model types generalize in a way more consistent with anincorrect linear rule than the correct hierarchical rule. These results suggestthat human-like generalization from text alone requires stronger biases thanthe general sequence-processing biases of standard neural networkarchitectures.

How poor is the stimulus? Evaluating hierarchical generalization in neural networks trained on child-directed speech