@article{sherborne-lapata-2023-meta,
title = "Meta-Learning a Cross-lingual Manifold for Semantic Parsing",
author = "Sherborne, Tom and
Lapata, Mirella",
journal = "Transactions of the Association for Computational Linguistics",
volume = "11",
year = "2023",
address = "Cambridge, MA",
publisher = "MIT Press",
url = "https://aclanthology.org/2023.tacl-1.4",
doi = "10.1162/tacl_a_00533",
pages = "49--67",
abstract = "Localizing a semantic parser to support new languages requires effective cross-lingual generalization. Recent work has found success with machine-translation or zero-shot methods, although these approaches can struggle to model how native speakers ask questions. We consider how to effectively leverage minimal annotated examples in new languages for few-shot cross-lingual semantic parsing. We introduce a first-order meta-learning algorithm to train a semantic parser with maximal sample efficiency during cross-lingual transfer. Our algorithm uses high-resource languages to train the parser and simultaneously optimizes for cross-lingual generalization to lower-resource languages. Results across six languages on ATIS demonstrate that our combination of generalization steps yields accurate semantic parsers sampling {\mbox{$\leq$}}10{\%} of source training data in each new language. Our approach also trains a competitive model on Spider using English with generalization to Chinese similarly sampling {\mbox{$\leq$}}10{\%} of training data.1",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sherborne-lapata-2023-meta">
<titleInfo>
<title>Meta-Learning a Cross-lingual Manifold for Semantic Parsing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tom</namePart>
<namePart type="family">Sherborne</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mirella</namePart>
<namePart type="family">Lapata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<genre authority="bibutilsgt">journal article</genre>
<relatedItem type="host">
<titleInfo>
<title>Transactions of the Association for Computational Linguistics</title>
</titleInfo>
<originInfo>
<issuance>continuing</issuance>
<publisher>MIT Press</publisher>
<place>
<placeTerm type="text">Cambridge, MA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">periodical</genre>
<genre authority="bibutilsgt">academic journal</genre>
</relatedItem>
<abstract>Localizing a semantic parser to support new languages requires effective cross-lingual generalization. Recent work has found success with machine-translation or zero-shot methods, although these approaches can struggle to model how native speakers ask questions. We consider how to effectively leverage minimal annotated examples in new languages for few-shot cross-lingual semantic parsing. We introduce a first-order meta-learning algorithm to train a semantic parser with maximal sample efficiency during cross-lingual transfer. Our algorithm uses high-resource languages to train the parser and simultaneously optimizes for cross-lingual generalization to lower-resource languages. Results across six languages on ATIS demonstrate that our combination of generalization steps yields accurate semantic parsers sampling łeq10% of source training data in each new language. Our approach also trains a competitive model on Spider using English with generalization to Chinese similarly sampling łeq10% of training data.1</abstract>
<identifier type="citekey">sherborne-lapata-2023-meta</identifier>
<identifier type="doi">10.1162/tacl_a_00533</identifier>
<location>
<url>https://aclanthology.org/2023.tacl-1.4</url>
</location>
<part>
<date>2023</date>
<detail type="volume"><number>11</number></detail>
<extent unit="page">
<start>49</start>
<end>67</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Journal Article
%T Meta-Learning a Cross-lingual Manifold for Semantic Parsing
%A Sherborne, Tom
%A Lapata, Mirella
%J Transactions of the Association for Computational Linguistics
%D 2023
%V 11
%I MIT Press
%C Cambridge, MA
%F sherborne-lapata-2023-meta
%X Localizing a semantic parser to support new languages requires effective cross-lingual generalization. Recent work has found success with machine-translation or zero-shot methods, although these approaches can struggle to model how native speakers ask questions. We consider how to effectively leverage minimal annotated examples in new languages for few-shot cross-lingual semantic parsing. We introduce a first-order meta-learning algorithm to train a semantic parser with maximal sample efficiency during cross-lingual transfer. Our algorithm uses high-resource languages to train the parser and simultaneously optimizes for cross-lingual generalization to lower-resource languages. Results across six languages on ATIS demonstrate that our combination of generalization steps yields accurate semantic parsers sampling łeq10% of source training data in each new language. Our approach also trains a competitive model on Spider using English with generalization to Chinese similarly sampling łeq10% of training data.1
%R 10.1162/tacl_a_00533
%U https://aclanthology.org/2023.tacl-1.4
%U https://doi.org/10.1162/tacl_a_00533
%P 49-67
Markdown (Informal)
[Meta-Learning a Cross-lingual Manifold for Semantic Parsing](https://aclanthology.org/2023.tacl-1.4) (Sherborne & Lapata, TACL 2023)
ACL