[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84608-en":3,"doc-seo-84608-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},84608,16904993612988,"Olivia Brown","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","Generative Retrieval for Table Union Search","Modern data lakes store heterogeneous tables where task-relevant information is spread across different schemas, sources, and naming conventions. Table union search (TUS) retrieves tables that can be reliably unioned with a query table, enabling discovery, enrichment, and downstream analytics. Learning-based TUS still relies on an encode–search–refine pipeline, so quality depends on candidate-pool recall and incurs rising latency and storage costs at scale. GenTUS reformulates TUS as constrained generation over discrete semantic table identifiers, ranking unionable tables directly.","Generative Retrieval for Table Union Search  \nShulun Zhang  \nThe Chinese University of Hong Kong, Shenzhen Shenzhen, China [shulunzhang@link.cuhk.edu.cn](shulunzhang@link.cuhk.edu.cn)  \nLinting Wang  \nFudan University Shanghai, China [lintingwang25@m.fudan.edu.cn](lintingwang25@m.fudan.edu.cn)  \nYuwei Xu  \nThe Chinese University of Hong Kong, Shenzhen Shenzhen, China [yuweixu@link.cuhk.edu.cn](yuweixu@link.cuhk.edu.cn)  \narXiv :2607 .00833v 1 [ cs .DB] 1 Jul 2026  \nYingli Zhou  \nThe Chinese University of Hong Kong, Shenzhen Shenzhen, China [yinglizhou@link.cuhk.edu.cn](yinglizhou@link.cuhk.edu.cn)  \nChenhao Ma  \nThe Chinese University of Hong Kong, Shenzhen Shenzhen, China [machenhao@cuhk.edu.cn](machenhao@cuhk.edu.cn)  \nAbstract  \nModern data lakes contain heterogeneous tables whose task-relevant information is often scattered across different schemas, sources, and naming conventions. Table union search (TUS) retrieves tables that can be reliably unioned with a query table, supporting data discovery, enrichment, and downstream analytics. Although learning-based TUS methods improve table-or column-level representations, they still follow an encode–search–refine pipeline: candidate retrieval is followed by query–candidate matching or reranking, making quality dependent on candidate-pool recall and incurring growing latency and storage costs as the data lake scales. We propose GenTUS, a generative retrieval framework that reformulates TUS as constrained generation over discrete semantic table identifiers. Instead of searching and reranking an explicit candidate pool, GenTUS assigns candidate tables compact unionabilityaware identifiers and trains a generator to produce the identifiers of unionable tables directly from the query. At query time, constrained decoding ensures that generated identifiers correspond to valid data-lake tables and returns them as ranked retrieval results. Experiments on seven public TUS benchmarks show that GenTUS achieves the best overall retrieval quality, with an average rank of 1.05 compared to 2.57 for the strongest baseline, while substantially reducing online latency, retrieval-artifact storage, and incremental update cost.  \nReference Format:  \nShulun Zhang, Linting Wang, Yuwei Xu, Yingli Zhou, and Chenhao Ma. Generative Retrieval for Table Union Search.  \nPVLDB Artifact Availability:  \nThe source code, data, and/or other artifacts have been made available at [https://github.com/alanzhang1001/GenTUS](https://github.com/alanzhang1001/GenTUS).  \nThis work is licensed under the Creative Commons BY-NC-ND 4.0 International License. Visit [https://creativecommons.org/licenses/by-nc-nd/4.0/ to view a copy of](https://creativecommons.org/licenses/by-nc-nd/4.0/ to view a copy of)[ ](https://creativecommons.org/licenses/by-nc-nd/4.0/ to view a copy of)[this license. For any use beyond those covered by this license](this license. For any use beyond those covered by this license), [obtain permission by](obtain permission by)[emailing info@vldb.org. Copyright](emailing info@vldb.org. Copyright) is held by the owner/author(s). Publication rights licensed to the VLDB Endowment.  \nProceedings of the VLDB Endowment, Vol. XX, No. XX ISSN 2150-8097 . doi:XX.XX/XXX.XX  \nFigure 1: An example of table union search.  \n1 Introduction  \nModern data lakes contain large collections of heterogeneous tables published by different organizations and curated under different schemas, metadata conventions, and quality standards [1, 9, 18, 32, 49] . To support downstream analysis, users often need to discover tables from a data lake that are relevant to a given analytical task. In many cases, the information needed to answer a given task isnot always contained in a single table; instead, it is scattered across multiple tables with overlapping semantics but different schemas, naming conventions, and coverage [14, 32, 51] . To guide such discovery, a query table, which may come from an existing dataset, an intermediate analysis resu","cbCaipPFsciS9sqW","https://ap.wps.com/l/cbCaipPFsciS9sqW","pdf",1666342,1,14,"English","en",105,"# Introduction\n## Table union search (TUS)\n## Generative retrieval approach","[{\"question\":\"What problem does table union search (TUS) address?\",\"answer\":\"TUS retrieves tables from a data lake that can be reliably unioned with a given query table, aligning columns so rows can be appended based on semantic compatibility rather than just keyword overlap or shared keys.\"},{\"question\":\"Why do existing learning-based TUS methods face scalability issues?\",\"answer\":\"They typically follow an encode–search–refine pipeline where retrieval quality depends on candidate-pool recall, and increasing dataset scale leads to higher latency and storage costs for candidate indexing and refinement.\"},{\"question\":\"How does GenTUS reformulate TUS compared with prior candidate retrieval methods?\",\"answer\":\"GenTUS turns TUS into constrained generation: it assigns unionability-aware compact semantic table identifiers to candidate tables and trains a generator to produce identifiers of unionable tables directly from the query, using constrained decoding to ensure validity.\"}]",1784197084,35,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"generative-retrieval-for-table-union-search","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/generative-retrieval-for-table-union-search/84608/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does table union search (TUS) address?","Question",{"text":75,"@type":76},"TUS retrieves tables from a data lake that can be reliably unioned with a given query table, aligning columns so rows can be appended based on semantic compatibility rather than just keyword overlap or shared keys.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Why do existing learning-based TUS methods face scalability issues?",{"text":80,"@type":76},"They typically follow an encode–search–refine pipeline where retrieval quality depends on candidate-pool recall, and increasing dataset scale leads to higher latency and storage costs for candidate indexing and refinement.",{"name":82,"@type":73,"acceptedAnswer":83},"How does GenTUS reformulate TUS compared with prior candidate retrieval methods?",{"text":84,"@type":76},"GenTUS turns TUS into constrained generation: it assigns unionability-aware compact semantic table identifiers to candidate tables and trains a generator to produce identifiers of unionable tables directly from the query, using constrained decoding to ensure validity.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]