Skip to main navigation Skip to search Skip to main content

SIGMORPHON-UniMorph 2022 Shared Task 0: Modeling Inflection in Language Acquisition

  • Stony Brook University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

12 Scopus citations

Abstract

This year's iteration of the SIGMORPHON-UniMorph shared task on “human-like” morphological inflection generation focuses on generalization and errors in language acquisition. Systems are trained on data sets extracted from corpora of child-directed speech in order to simulate a natural learning setting, and their predictions are evaluated against what is known about children's developmental trajectories for three well-studied patterns: English past tense, German noun plurals, and Arabic noun plurals. Three submitted neural systems were evaluated together with two baselines. Performance was generally good, and all systems were prone to human-like over-regularization. However, all systems were also prone to non-human-like over-irregularization and nonsense productions to varying degrees. We situate this behavior in a discussion of the Past Tense Debate.

Original languageEnglish
Title of host publicationSIGMORPHON 2022 - 19th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, Proceedings of the Workshop
EditorsGarrett Nicolai, Eleanor Chodroff
PublisherAssociation for Computational Linguistics (ACL)
Pages157-175
Number of pages19
ISBN (Electronic)9781955917827
DOIs
StatePublished - 2022
Event19th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, SIGMORPHON 2022 - Seattle, United States
Duration: Jul 14 2022 → …

Publication series

NameSIGMORPHON 2022 - 19th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, Proceedings of the Workshop

Conference

Conference19th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, SIGMORPHON 2022
Country/TerritoryUnited States
CitySeattle
Period07/14/22 → …

Fingerprint

Dive into the research topics of 'SIGMORPHON-UniMorph 2022 Shared Task 0: Modeling Inflection in Language Acquisition'. Together they form a unique fingerprint.

Cite this