[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83184-en":3,"doc-seo-83184-105":29,"detail-sidebar-cat-0-en-105":83},{"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},83184,1374391974468,"Eden","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","ShapeTalk: Combining Natural Language and Sketch for Time-Series Pattern Querying","Searching for time-series segments that match user-defined patterns is important in finance, climate science, and healthcare, yet existing visual query tools struggle with vague, composite, or fuzzy descriptions and often demand precise sketches or rigid structured filters. ShapeTalk introduces a coordinated natural-language and sketch-based system for univariate time-series pattern search. Text captures semantic and compositional intent, while sketching enables geometric refinement. Both modalities stay synchronized via shared visual context and editable shape-feature constraints implemented with an LLM semantic parsing pipeline, supported by usage scenarios and evaluation results.","arXiv :2607 .07073v 1 [ cs .HC] 8 Jul 2026  \nShapeTalk: Combining Natural Language and Sketch for Time-Series Pattern Querying  \nGuoruizhe Sun* , Yueqiao Chen* , Emily Guo, Yutong Yao, and Dongyu Liu   \nFig. 1: User interface of ShapeTalk, shown with the Sacramento Weather dataset from 2006 to 2008 . Control Panel (A), where users import a dataset and set the query window length. Chat View (B), which supports natural-language queries via an LLM; B1 is the input box for user queries. Query View by Text (C), which shows results for natural-language and feature-based queries; C1 is the Feature Panel for adjusting feature selections and C2 displays snapshots of matched time-series windows. Query View by Sketch (D), which supports sketch-based queries; D1 is the sketch canvas and D2 shows matched windows for the sketch. Full Time Series View (E), which provides an overview of the entire time series and highlights matches from each query modality. F1 presents the parameter-adjustment view for global feature, while F2 is for local feature. The figure also illustrates a typical exploration loop for finding suitable weather to visit Sacramento: 1. sketch to obtain initial matches, 2. ask the LLM to describe the sketch, and 3&4. refine queries using the resulting natural-language description.  \nAbstract—Searching for time-series segments that match user-defined patterns is important in domains such as finance, climate science, and healthcare. However, existing visual query tools often struggle to support vague, composite, or fuzzy pattern descriptions, often requiring users to express their intent through precise sketches or rigid structured filters. We present ShapeTalk, a coordinated natural-language and sketch-based querying system for univariate time-series pattern search. Rather than treating text and sketch as a fused input stream, ShapeTalk uses them as complementary representations of analytic intent: natural language supports semantic and compositional pattern descriptions, while sketching supports direct geometric refinement. The two modalities are linked through a shared visual context, editable feature representations, and synchronized result views, enabling users to move between text and sketch during iterative query formulation. At its core is an LLM-based semantic parsing pipeline that translates free-form natural-language queries into interpretable and editable shape-feature constraints. We evaluate ShapeTalk through two usage scenarios, a user study with failure-case analysis, and an assessment of the LLM-based semantic parsing pipeline. The results show that ShapeTalk supports effective time-series pattern search, with natural language serving as an accessible entry point and sketching providing a complementary mechanism for refinement and recovery when textual specifications are insufficient.  \nIndex Terms—Time Series, Pattern Query, Natural Language, Sketch, Visual Analytics  \n• All authors are with the University of California at Davis. E-mail: {grzsun, yeqchen, emiguo, ytyao, [dyuliu}@ucdavis.edu](dyuliu}@ucdavis.edu);  \n• * Guoruizhe Sun and Yueqiao Chen contributed equally to this work. Manuscript received xx xxx. 202x; accepted xx xxx. 202x. Date of Publication xx xxx. 202x; date of current version xx xxx. 202x. For information on obtaining reprints of this article, please send e-mail to: [reprints@ieee.org](reprints@ieee.org). Digital Object Identifier: xx.xxxx/TVCG.202x.xxxxxxx  \n1 INTRODUCTION  \nTime-series query refers to retrieving segments that match a userdefined temporal pattern and is a fundamental task in many domains. Analysts often search large time-series collections for specific patterns or anomalies in finance [16], transportation [29], climate science [35], healthcare [5], and beyond. For example, an investment analyst may ask for “stocks that tanked in 2020,” seeking trajectories that exhibit a sharp drop during that period. Broadly, analysts across domains rely on  \npattern search to identify s","cbCaiiI71k3xpJZq","https://ap.wps.com/l/cbCaiiI71k3xpJZq","pdf",7886010,1,32,"English","en",105,"# Introduction\n## Time-series query and related domains\n## Limitations of existing visual pattern search\n## Challenges of natural-language interpretation\n## Controlled vocabulary approaches","[{\"question\":\"What is the role of the LLM in ShapeTalk?\",\"answer\":\"An LLM-based semantic parsing pipeline translates free-form natural-language queries into interpretable and editable shape-feature constraints, enabling iterative query formulation and alignment with sketch-derived refinement.\"}]",1784185822,81,{"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":78,"head_meta":80,"extra_data":82,"updated_unix":27},"shapetalk-combining-natural-language-and-sketch-for-time-series-pattern-querying","",{"@graph":35,"@context":77},[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/shapetalk-combining-natural-language-and-sketch-for-time-series-pattern-querying/83184/",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],{"name":72,"@type":73,"acceptedAnswer":74},"What is the role of the LLM in ShapeTalk?","Question",{"text":75,"@type":76},"An LLM-based semantic parsing pipeline translates free-form natural-language queries into interpretable and editable shape-feature constraints, enabling iterative query formulation and alignment with sketch-derived refinement.","Answer","https://schema.org",{"og:url":51,"og:type":79,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":81,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":84},[85,89,93,97,102,107,112,115,120,123,127],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":86,"show_sort_weight":87,"slug":88},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":90,"show_sort_weight":91,"slug":92},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Exam",70,"exam",{"id":98,"doc_module":4,"doc_module_name":45,"category_name":99,"show_sort_weight":100,"slug":101},5,"Comic",60,"comic",{"id":103,"doc_module":4,"doc_module_name":45,"category_name":104,"show_sort_weight":105,"slug":106},6,"Technology",50,"technology",{"id":108,"doc_module":4,"doc_module_name":45,"category_name":109,"show_sort_weight":110,"slug":111},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":113,"slug":114},30,"research-report",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":45,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":98,"slug":130},19,"General","general"]