[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-273573-105":59,"doc-detail-273573-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","omnialign-a-unified-multilingual-aligner-for-word-and-sentence-alignment","OmniAlign - A Unified Multilingual Aligner for Word and Sentence Alignment","","Cross-lingual sequence alignment is central to building and using parallel corpora, from document and sentence correspondences down to word and subword mappings. Existing tools usually target only one granularity, forcing practitioners to maintain separate pipelines for word- and sentence-level alignment, particularly in multilingual and long-text cases. OmniAlign provides a unified multilingual solution supporting both word- and sentence-level alignment with one lightweight model. It uses a long-context encoder, derives word alignments from token similarity matrices, and computes m–n sentence alignments via sentence embeddings with dynamic programming. A four-stage training pipeline balances accuracy and representation quality. Experiments show strong results on alignment benchmarks and robust generalization to unseen language pairs. Further supervised fine-tuning on short texts improves alignment quality while retaining long-context capability for long-text word alignment.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":30,"@type":76,"position":81},"https://docshare.wps.com/document/technology/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/omnialign-a-unified-multilingual-aligner-for-word-and-sentence-alignment/273573/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/omnialign-a-unified-multilingual-aligner-for-word-and-sentence-alignment/273573.png","ImageObject",300,407,{"name":92,"@type":93},"Liam","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-19","2026-09-15",true,{"@type":102,"interactionType":103,"userInteractionCount":44},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does OmniAlign address in cross-lingual sequence alignment?","Question",{"text":112,"@type":113},"It addresses the lack of a single practical aligner that supports both word-level and sentence-level alignment across languages, especially under multilingual and long-text settings.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does OmniAlign produce word-level alignments?",{"text":117,"@type":113},"It induces word alignments from contextualized token similarity matrices generated by a long-context encoder.",{"name":119,"@type":110,"acceptedAnswer":120},"How are sentence-level m–n alignments computed?",{"text":121,"@type":113},"It obtains sentence embeddings and then determines document-level m–n sentence alignments using dynamic programming based on similarity.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},273573,1791009697,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":29,"category_name":30,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":44,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":144,"read_time":145},8796095461564,"https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d","arXiv : 2608 . 18474v 1 [cs .CL] 19 Aug 2026  \nOmniAlign: A Unified Multilingual Aligner for Word and  \nSentence Alignment  \nMengpeng Yang Jingxu Yang Chao Chen Tian Xia Yabo Sun Qiang Liu  \nWPS Qingqiu / Wuhan, China  \n[yangmengpeng@wps.cn](yangmengpeng@wps.cn) , [yangjingxu1@wps.cn](yangjingxu1@wps.cn) , [chenchao9@wps.cn](chenchao9@wps.cn)  \n[xiatian3@wps.cn](xiatian3@wps.cn) , [sunyabo@wps.cn](sunyabo@wps.cn) , [liuqiang2@wps.cn](liuqiang2@wps.cn)  \nAbstract  \nCross-lingual sequence alignment is fundamental for building and exploiting parallel corpora, spanning mappings from documents and sentences down to words and subwords.  \nExisting tools, however, typically specialize in a single granularity, so practitioners often need separate systems for word-and sentence-level alignment—especially in multilingual and long-text settings. We present OmniAlign, a unified multilingual aligner that supports both word-level and sentence-level alignment with a single lightweight model. Built on an encoder-only backbone with strong long-context modeling, OmniAlign induces word alignments from contextualized token similarity matrices, and obtains document-level m–n sentence alignments via sentence embeddings combined with dynamic programming. To balance fine-grained alignment accuracy and sentencerepresentation quality, we use a four-stage training pipeline: alignment-oriented continued pre-training, self-supervised learning, supervised fine-tuning on human annotations, and sentence-embedding distillation from a strong multilingual teacher. Experiments show that OmniAlign achieves highly competitive performance on both word-and sentence-alignment benchmarks and generalizes well to unseen language pairs. Surprisingly, later-stage supervised fine-tuning on short texts further improves alignment quality while retaining the long-context understanding acquired in earlier training, keeping the model robust on long-text word alignment.  \nCode: [https://github.com/MilkDargon/OmniAlign](https://github.com/MilkDargon/OmniAlign)  \nModel: [https://huggingface.co/WPS-Qingqiu/OmniAlign](https://huggingface.co/WPS-Qingqiu/OmniAlign)  \n1 Introduction  \nCross-lingual Sequence Alignment is a core foundational task in the construction and exploitation of parallel corpora. Its goal is to establish precise and consistent correspondences between text sequences across different languages. This task typically spans multiple levels of granularity—from coarse-grained alignment at the document or paragraph level, to sentence-level alignment, and further down to finegrained word-or subword-level alignment. Together, these alignment processes form the critical pathway for cross-lingual information mapping, enabling models to build stable semantic correspondences in multilingual spaces. They provide essential support for tasks such as building para-crawl pipelines, constructing translation memories in computer-assisted translation systems (Varga et al., 2008), projecting linguistic annotations and XML structural labels across languages (David et al., 2001; Hashimoto et al., 2019), detecting under-translation in neural machine translation (Tu et al., 2016), developing cross-lingual transfer NLP tools (Nicolai & Yarowsky, 2019; Mayhew et al., 2017), and improving or augmenting multilingual pre-trained models (Chi et al., 2021) .  \nDespite the emergence of various models targeting sequence alignment at different granularities in recent years, there is still a lack of a single practical aligner that covers both word-and sentence-level needs while remaining multilingual and long-context capable. Previous studies typically focus on only one subtask of the alignment problem. For example, methods such as (Sabet et al., 2020; Dou & Neubig, 2021; Wang et al., 2022; Wu et al., 2023; Nagata et al., 2020; Lai et al., 2022; Latouche et al., 2024) primarily explore word alignment; however, most of them suffer from significant performance degradation as sequence length increases and a","cbCaiqIlH6c6jtZ0","https://ap.wps.com/l/cbCaiqIlH6c6jtZ0","pdf",896308,15,"English","# Introduction\n## Problem and Motivation\n## OmniAlign Approach\n## Training Pipeline\n## Experiments and Results","[{\"question\":\"What problem does OmniAlign address in cross-lingual sequence alignment?\",\"answer\":\"It addresses the lack of a single practical aligner that supports both word-level and sentence-level alignment across languages, especially under multilingual and long-text settings.\"},{\"question\":\"How does OmniAlign produce word-level alignments?\",\"answer\":\"It induces word alignments from contextualized token similarity matrices generated by a long-context encoder.\"},{\"question\":\"How are sentence-level m–n alignments computed?\",\"answer\":\"It obtains sentence embeddings and then determines document-level m–n sentence alignments using dynamic programming based on similarity.\"}]","OmniAlign - A Unified Multilingual Aligner for Word and Sentence Alignment | PDF",1789472122,38]