LKMT: Linguistics Knowledge-DrivenMulti-Task NeuralMachine Translation for Urdu and English

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F25%3AMHAJJG69" target="_blank" >RIV/00216208:11320/25:MHAJJG69 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.scopus.com/inward/record.uri?eid=2-s2.0-85206828110&doi=10.32604%2fcmc.2024.054673&partnerID=40&md5=11cc8e5a13114b45ab4a4331dd861ce1" target="_blank" >https://www.scopus.com/inward/record.uri?eid=2-s2.0-85206828110&doi=10.32604%2fcmc.2024.054673&partnerID=40&md5=11cc8e5a13114b45ab4a4331dd861ce1</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.32604/cmc.2024.054673" target="_blank" >10.32604/cmc.2024.054673</a>

Jazyk výsledku

angličtina

Název v původním jazyce

LKMT: Linguistics Knowledge-DrivenMulti-Task NeuralMachine Translation for Urdu and English

Popis výsledku v původním jazyce

Thanks to the strong representation capability of pre-trained language models, supervised machine translation models have achieved outstanding performance. However, the performances of these models drop sharply when the scale of the parallel training corpus is limited. Considering the pre-trained language model has a strong ability formonolingual representation, it is the key challenge formachine translation to construct the in-depth relationship between the source and target language by injecting the lexical and syntactic information into pre-trained language models. To alleviate the dependence on the parallel corpus, we propose a Linguistics Knowledge-Driven Multi- Task (LKMT) approach to inject part-of-speech and syntactic knowledge into pre-trained models, thus enhancing the machine translation performance. On the one hand, we integrate part-of-speech and dependency labels into the embedding layer and exploit large-scale monolingual corpus to update all parameters of pre-trained language models, thus ensuring the updated language model contains potential lexical and syntactic information. On the other hand, we leverage an extra self-attention layer to explicitly inject linguistic knowledge into the pre-trained language model-enhanced machine translation model. Experiments on the benchmark dataset show that our proposed LKMT approach improves the Urdu-English translation accuracy by 1.97 points and the English-Urdu translation accuracy by 2.42 points, highlighting the effectiveness of our LKMT framework. Detailed ablation experiments confirm the positive impact of part-of-speech and dependency parsing on machine translation. © 2024 The Authors.

Název v anglickém jazyce

LKMT: Linguistics Knowledge-DrivenMulti-Task NeuralMachine Translation for Urdu and English

Popis výsledku anglicky

Thanks to the strong representation capability of pre-trained language models, supervised machine translation models have achieved outstanding performance. However, the performances of these models drop sharply when the scale of the parallel training corpus is limited. Considering the pre-trained language model has a strong ability formonolingual representation, it is the key challenge formachine translation to construct the in-depth relationship between the source and target language by injecting the lexical and syntactic information into pre-trained language models. To alleviate the dependence on the parallel corpus, we propose a Linguistics Knowledge-Driven Multi- Task (LKMT) approach to inject part-of-speech and syntactic knowledge into pre-trained models, thus enhancing the machine translation performance. On the one hand, we integrate part-of-speech and dependency labels into the embedding layer and exploit large-scale monolingual corpus to update all parameters of pre-trained language models, thus ensuring the updated language model contains potential lexical and syntactic information. On the other hand, we leverage an extra self-attention layer to explicitly inject linguistic knowledge into the pre-trained language model-enhanced machine translation model. Experiments on the benchmark dataset show that our proposed LKMT approach improves the Urdu-English translation accuracy by 1.97 points and the English-Urdu translation accuracy by 2.42 points, highlighting the effectiveness of our LKMT framework. Detailed ablation experiments confirm the positive impact of part-of-speech and dependency parsing on machine translation. © 2024 The Authors.

Druh

J_SC - Článek v periodiku v databázi SCOPUS

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 periodika

Computers, Materials and Continua

ISSN

1546-2218

e-ISSN

—

Svazek periodika

81

Číslo periodika v rámci svazku

1

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

19

Strana od-do

951-969

Kód UT WoS článku

—

EID výsledku v databázi Scopus

2-s2.0-85206828110