Saved in:
Bibliographic Details
Main Authors: Mahon, Louis, Johnson, Mark, Steedman, Mark
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.12832
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909539109961728
author Mahon, Louis
Johnson, Mark
Steedman, Mark
author_facet Mahon, Louis
Johnson, Mark
Steedman, Mark
contents This work develops a probabilistic child language acquisition model to learn a range of linguistic phenonmena, most notably long-range syntactic dependencies of the sort found in object wh-questions, among other constructions. The model is trained on a corpus of real child-directed speech, where each utterance is paired with a logical form as a meaning representation. It then learns both word meanings and language-specific syntax simultaneously. After training, the model can deduce the correct parse tree and word meanings for a given utterance-meaning pair, and can infer the meaning if given only the utterance. The successful modelling of long-range dependencies is theoretically important because it exploits aspects of the model that are, in general, trans-context-free.
format Preprint
id arxiv_https___arxiv_org_abs_2503_12832
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Modelling Child Learning and Parsing of Long-range Syntactic Dependencies
Mahon, Louis
Johnson, Mark
Steedman, Mark
Computation and Language
This work develops a probabilistic child language acquisition model to learn a range of linguistic phenonmena, most notably long-range syntactic dependencies of the sort found in object wh-questions, among other constructions. The model is trained on a corpus of real child-directed speech, where each utterance is paired with a logical form as a meaning representation. It then learns both word meanings and language-specific syntax simultaneously. After training, the model can deduce the correct parse tree and word meanings for a given utterance-meaning pair, and can infer the meaning if given only the utterance. The successful modelling of long-range dependencies is theoretically important because it exploits aspects of the model that are, in general, trans-context-free.
title Modelling Child Learning and Parsing of Long-range Syntactic Dependencies
topic Computation and Language
url https://arxiv.org/abs/2503.12832