Salvato in:
Dettagli Bibliografici
Autori principali: Zhang, Jingshen, Qiu, Xinying, Shen, Teng, Wang, Wenyu, Zhang, Kailin, Feng, Wenhe
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2407.05054
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914859987238912
author Zhang, Jingshen
Qiu, Xinying
Shen, Teng
Wang, Wenyu
Zhang, Kailin
Feng, Wenhe
author_facet Zhang, Jingshen
Qiu, Xinying
Shen, Teng
Wang, Wenyu
Zhang, Kailin
Feng, Wenhe
contents Cross-lingual word alignment plays a crucial role in various natural language processing tasks, particularly for low-resource languages. Recent study proposes a BiLSTM-based encoder-decoder model that outperforms pre-trained language models in low-resource settings. However, their model only considers the similarity of word embedding spaces and does not explicitly model the differences between word embeddings. To address this limitation, we propose incorporating contrastive learning into the BiLSTM-based encoder-decoder framework. Our approach introduces a multi-view negative sampling strategy to learn the differences between word pairs in the shared cross-lingual embedding space. We evaluate our model on five bilingual aligned datasets spanning four ASEAN languages: Lao, Vietnamese, Thai, and Indonesian. Experimental results demonstrate that integrating contrastive learning consistently improves word alignment accuracy across all datasets, confirming the effectiveness of the proposed method in low-resource scenarios. We will release our data set and code to support future research on ASEAN or more low-resource word alignment.
format Preprint
id arxiv_https___arxiv_org_abs_2407_05054
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Cross-Lingual Word Alignment for ASEAN Languages with Contrastive Learning
Zhang, Jingshen
Qiu, Xinying
Shen, Teng
Wang, Wenyu
Zhang, Kailin
Feng, Wenhe
Computation and Language
Cross-lingual word alignment plays a crucial role in various natural language processing tasks, particularly for low-resource languages. Recent study proposes a BiLSTM-based encoder-decoder model that outperforms pre-trained language models in low-resource settings. However, their model only considers the similarity of word embedding spaces and does not explicitly model the differences between word embeddings. To address this limitation, we propose incorporating contrastive learning into the BiLSTM-based encoder-decoder framework. Our approach introduces a multi-view negative sampling strategy to learn the differences between word pairs in the shared cross-lingual embedding space. We evaluate our model on five bilingual aligned datasets spanning four ASEAN languages: Lao, Vietnamese, Thai, and Indonesian. Experimental results demonstrate that integrating contrastive learning consistently improves word alignment accuracy across all datasets, confirming the effectiveness of the proposed method in low-resource scenarios. We will release our data set and code to support future research on ASEAN or more low-resource word alignment.
title Cross-Lingual Word Alignment for ASEAN Languages with Contrastive Learning
topic Computation and Language
url https://arxiv.org/abs/2407.05054