Enhancing Few-shot Cross-lingual Transfer with Target Language Peculiar Examples

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F23%3AGCFGY9IY" target="_blank" >RIV/00216208:11320/23:GCFGY9IY - isvavai.cz</a>

Result on the web

<a href="https://aclanthology.org/2023.findings-acl.47/" target="_blank" >https://aclanthology.org/2023.findings-acl.47/</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.18653/v1/2023.findings-acl.47" target="_blank" >10.18653/v1/2023.findings-acl.47</a>

Result language

angličtina

Original language name

Enhancing Few-shot Cross-lingual Transfer with Target Language Peculiar Examples

Original language description

"Few-shot cross-lingual transfer, fine-tuning Multilingual Masked Language Model (MMLM) with source language labeled data and a small amount of target language labeled data, provides excellent performance in the target language. However, if no labeled data in the target language are available, they need to be created through human annotations. In this study, we devise a metric to select annotation candidates from an unlabeled data pool that efficiently enhance accuracy for few-shot cross-lingual transfer."

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

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

Project

—

Continuities

—

Publication year

2023

Confidentiality

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

Article name in the collection

"Findings of the Association for Computational Linguistics: ACL 2023"

ISBN

978-1-959429-62-3

ISSN

—

e-ISSN

—

Number of pages

21

Pages from-to

747-767

Publisher name

ACL

Place of publication

Toronto, Canada

Event location

Toronto, Canada

Event date

Jan 1, 2023

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—