Transferability of Syntax-Aware Graph Neural Networks in Zero-Shot Cross-Lingual Semantic Role Labeling

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F25%3ACKW6C8UD" target="_blank" >RIV/00216208:11320/25:CKW6C8UD - isvavai.cz</a>

Výsledek na webu

<a href="https://www.scopus.com/inward/record.uri?eid=2-s2.0-85217623093&partnerID=40&md5=5f7ed0d162a62107b223ccca9faac10a" target="_blank" >https://www.scopus.com/inward/record.uri?eid=2-s2.0-85217623093&partnerID=40&md5=5f7ed0d162a62107b223ccca9faac10a</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Transferability of Syntax-Aware Graph Neural Networks in Zero-Shot Cross-Lingual Semantic Role Labeling

Popis výsledku v původním jazyce

Recent models in cross-lingual semantic role labeling (SRL) barely analyze the applicability of their network selection. We believe that network selection is important since it affects the transferability of cross-lingual models, i.e., how the model can extract universal features from source languages to label target languages. Therefore, we comprehensively compare the transferability of different graph neural network (GNN)-based models enriched with universal dependency trees. GNN-based models include transformer-based, graph convolutional network-based, and graph attention network (GAT)-based models. We focus our study on a zero-shot setting by training the models in English and evaluating the models in 23 target languages provided by the Universal Proposition Bank. Based on our experiments, we consistently show that syntax from universal dependency trees is essential for cross-lingual SRL models to achieve better transferability. Dependency-aware self-attention with relative position representations (SAN-RPRs) transfer best across languages, especially in the long-range dependency distance. We also show that dependency-aware two-attention relational GATs transfer better than SAN-RPRs in languages where most arguments lie in a 1-2 dependency distance. © 2024 Association for Computational Linguistics.

Název v anglickém jazyce

Transferability of Syntax-Aware Graph Neural Networks in Zero-Shot Cross-Lingual Semantic Role Labeling

Popis výsledku anglicky

Recent models in cross-lingual semantic role labeling (SRL) barely analyze the applicability of their network selection. We believe that network selection is important since it affects the transferability of cross-lingual models, i.e., how the model can extract universal features from source languages to label target languages. Therefore, we comprehensively compare the transferability of different graph neural network (GNN)-based models enriched with universal dependency trees. GNN-based models include transformer-based, graph convolutional network-based, and graph attention network (GAT)-based models. We focus our study on a zero-shot setting by training the models in English and evaluating the models in 23 target languages provided by the Universal Proposition Bank. Based on our experiments, we consistently show that syntax from universal dependency trees is essential for cross-lingual SRL models to achieve better transferability. Dependency-aware self-attention with relative position representations (SAN-RPRs) transfer best across languages, especially in the long-range dependency distance. We also show that dependency-aware two-attention relational GATs transfer better than SAN-RPRs in languages where most arguments lie in a 1-2 dependency distance. © 2024 Association for Computational Linguistics.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

—

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

EMNLP - Conf. Empir. Methods Nat. Lang. Process., Find. EMNLP

ISBN

979-889176168-1

ISSN

—

e-ISSN

—

Počet stran výsledku

23

Strana od-do

20-42

Název nakladatele

Association for Computational Linguistics (ACL)

Místo vydání

—

Místo konání akce

Miami

Datum konání akce

1. 1. 2025

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—