Saved in:
Bibliographic Details
Main Authors: Huang, Baorong, Asiri, Ali
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.21842
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912799378112512
author Huang, Baorong
Asiri, Ali
author_facet Huang, Baorong
Asiri, Ali
contents High-quality parallel corpora are essential for Machine Translation (MT) research and translation teaching. However, Arabic-English resources remain scarce and existing datasets mainly consist of simple one-to-one mappings. In this paper, we present AlignAR, a generative sentence alignment method, and a new Arabic-English dataset comprising simple legal and complex literary parallel texts. Our evaluation demonstrates that "Easy" datasets lack the discriminatory power to fully assess alignment methods. By reducing one-to-one mappings in our "Hard" subset, we exposed the limitations of traditional alignment methods. In contrast, LLM-based approaches demonstrated better robustness, achieving an overall F1-score of 85.5%, a nearly 9% improvement over previous methods. Our datasets and codes are open-sourced at https://github.com/XXX.
format Preprint
id arxiv_https___arxiv_org_abs_2512_21842
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AlignAR: Generative Sentence Alignment for Arabic-English Parallel Corpora of Legal and Literary Texts
Huang, Baorong
Asiri, Ali
Computation and Language
High-quality parallel corpora are essential for Machine Translation (MT) research and translation teaching. However, Arabic-English resources remain scarce and existing datasets mainly consist of simple one-to-one mappings. In this paper, we present AlignAR, a generative sentence alignment method, and a new Arabic-English dataset comprising simple legal and complex literary parallel texts. Our evaluation demonstrates that "Easy" datasets lack the discriminatory power to fully assess alignment methods. By reducing one-to-one mappings in our "Hard" subset, we exposed the limitations of traditional alignment methods. In contrast, LLM-based approaches demonstrated better robustness, achieving an overall F1-score of 85.5%, a nearly 9% improvement over previous methods. Our datasets and codes are open-sourced at https://github.com/XXX.
title AlignAR: Generative Sentence Alignment for Arabic-English Parallel Corpora of Legal and Literary Texts
topic Computation and Language
url https://arxiv.org/abs/2512.21842