Please use this identifier to cite or link to this item: doi:10.22028/D291-36103
Volltext verfügbar? / Dokumentlieferung
Title: Compositional Generalization Requires Compositional Parsers
Author(s): Weißenhorn, Pia
Yao, Yuekun
Donatelli, Lucia
Koller, Alexander
Language: English
Publisher/Platform: arXiv
Year of Publication: 2022
Publikation type: Other
Abstract: A rapidly growing body of research on compositional generalization investigates the ability of a semantic parser to dynamically recombine linguistic elements seen in training into unseen sequences. We present a systematic comparison of sequence-to-sequence models and models guided by compositional principles on the recent COGS corpus (Kim and Linzen, 2020). Though seq2seq models can perform well on lexical tasks, they perform with near-zero accuracy on structural generalization tasks that require novel syntactic structures; this holds true even when they are trained to predict syntax instead of semantics. In contrast, compositional models achieve near-perfect accuracy on structural generalization; we present new results confirming this from the AM parser (Groschwitz et al., 2021). Our findings show structural generalization is a key measure of compositional generalization and requires models that are aware of complex structure.
DOI of the first publication: 10.48550/arXiv.2202.11937
URL of the first publication: https://arxiv.org/abs/2202.11937
Link to this record: hdl:20.500.11880/32885
http://dx.doi.org/10.22028/D291-36103
Date of registration: 5-May-2022
Notes: Preprint
Faculty: P - Philosophische Fakultät
Department: P - Sprachwissenschaft und Sprachtechnologie
Professorship: P - Prof. Dr. Alexander Koller
Collections:Die Universitätsbibliographie

Files for this record:
There are no files associated with this item.


Items in SciDok are protected by copyright, with all rights reserved, unless otherwise indicated.