[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83448-en":3,"doc-seo-83448-105":30,"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":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},83448,1099513958607,"Jiven","https://ap-avatar.wpscdn.com/avatar/100002390cf8733938c?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778829742770036399",8,"Research & Report","EVOTS: Evolutionary Transformer Search for Time Series Forecasting","Evolutionary neural architecture design for multivariate time-series forecasting is still limited, because most methods assume fixed Transformer structures even though tasks and forecasting settings vary greatly. This paper presents EVOTS, an evolutionary neural architecture search framework that discovers task-adaptive Transformer-like models via a modular genome encoding attention, feed-forward, and projection components. A repair mechanism preserves structural validity during evolution. Experiments on ETT (ETTh1/ETTh2/ETTm1/ETTm2) show competitive, sometimes improved mean squared error under multiple univariate and multivariate prediction settings, with additional training-time analysis.","arXiv :2607 .00154v1 [ cs .LG] 30 Jun 2026  \nEVOTS: EVOLUTIONARY TRANSFORMER SEARCH FOR TIME SERIES FORECASTING  \nAbdElRahman ElSaid Damir Pulatov  \n[elsaida@uncw.edu](elsaida@uncw.edu) [pulatovd@uncw.edu](pulatovd@uncw.edu)  \nUniversity of North Carolina Wilmington  \nWilmington, North Carolina, USA  \nJuly 2, 2026  \nABSTRACT  \nEvolutionary neural architecture design for multivariate time-series forecasting remains underexplored, with most approaches relying on fixed Transformer architectures despite substantial variation across tasks and forecasting settings. This paper introduces an evolutionary neural architecture search framework for discovering task-adaptive Transformer-like models for time-series forecasting (EVOTS) . Architectures are encoded using a modular genome representation that enables flexible composition of attention, feed-forward, and projection components, while a repair mechanism enforces structural validity throughout the evolutionary process. This formulation allows effective exploration of a diverse architecture space without relying on hand-crafted design rules.  \nThe proposed approach is evaluated on four benchmark datasets from the ETT family (ETTh1, ETTh2, ETTm1, and ETTm2) under multiple forecasting settings, including univariate-to-univariate, multivariate-to-univariate, and multivariate-to-multivariate prediction, with horizons of 96, 192, 336, and 720 . In the multivariate-to-multivariate setting, the evolved architectures achieve competitive and, in several cases, improved mean squared error relative to a strong Transformer-based baseline.  \nAdditional analyses examine performance differences across forecasting settings and report wall-clock training time to provide a coarse indication of computational cost.  \nOverall, the results demonstrate that evolutionary search can effectively discover flexible and highperforming Transformer-like architectures for multivariate time-series forecasting within practical runtime constraints.  \n1 Introduction  \nAccurate time-series forecasting is a critical component of many real-world systems, including energy management, industrial monitoring, and large-scale infrastructure control. Recent advances in Transformer-based architectures have significantly improved forecasting performance by enabling flexible modeling of long-range temporal dependencies and complex cross-variable interactions. As a result, Transformers and their variants have become a dominant paradigm in multivariate time-series forecasting.  \nDespite their success, most Transformer-based forecasting models rely on carefully hand-designed architectures. Choices such as tokenization strategy, attention formulation, block composition, and network depth are typically fixed a priori and tuned through extensive manual experimentation. While effective, this process is labor-intensive and often yields architectures that perform well on a narrow range of tasks or prediction horizons, limiting adaptability across datasets and forecasting regimes.  \nEvolutionary computation offers a principled alternative to manual architecture design. By treating neural architecturesas evolvable structures and optimizing them through population-based search, evolutionary methods can explore diverse architectural compositions without committing to a single design template. This paradigm is particularly well-suited to neural architecture search problems where the design space is highly structured and the cost of evaluation is substantial. In this work, we propose EVOTS1 , an evolutionary architecture search framework for Transformer-like time-series forecasting models. The framework represents architectures as modular genomes composed of Transformer-inspired  \n1[https://github.com/a-elsaid/EVOTS.git](https://github.com/a-elsaid/EVOTS.git)  \nA PREPRINT-JULY 2, 2026  \nprocessing blocks and evolves their composition using a steady-state evolutionary algorithm with weight inheritance. Unlike approaches that introduce new attentio","cbCaipnDKW4mkkTO","https://ap.wps.com/l/cbCaipnDKW4mkkTO","pdf",694298,4,1,13,"English","en",105,"# Introduction\n# Related Work","[{\"question\":\"What problem does EVOTS address in time-series forecasting?\",\"answer\":\"EVOTS targets the limitation that most Transformer-based forecasting methods rely on fixed, hand-designed architectures that may not adapt well across different tasks and forecasting regimes.\"},{\"question\":\"How does EVOTS search for Transformer-like architectures?\",\"answer\":\"EVOTS encodes architectures using a modular genome that composes attention, feed-forward, and projection components, then evolves them with a steady-state evolutionary algorithm with weight inheritance. A repair mechanism enforces structural validity during evolution.\"},{\"question\":\"What datasets and settings are used to evaluate the approach?\",\"answer\":\"The approach is evaluated on four ETT-family benchmark datasets (ETTh1, ETTh2, ETTm1, ETTm2) across multiple forecasting settings, including univariate-to-univariate, multivariate-to-univariate, and multivariate-to-multivariate prediction horizons of 96, 192, 336, and 720.\"}]",1784187955,33,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"evots-evolutionary-transformer-search-for-time-series-forecasting","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/evots-evolutionary-transformer-search-for-time-series-forecasting/83448/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-26","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 EVOTS address in time-series forecasting?","Question",{"text":75,"@type":76},"EVOTS targets the limitation that most Transformer-based forecasting methods rely on fixed, hand-designed architectures that may not adapt well across different tasks and forecasting regimes.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does EVOTS search for Transformer-like architectures?",{"text":80,"@type":76},"EVOTS encodes architectures using a modular genome that composes attention, feed-forward, and projection components, then evolves them with a steady-state evolutionary algorithm with weight inheritance. A repair mechanism enforces structural validity during evolution.",{"name":82,"@type":73,"acceptedAnswer":83},"What datasets and settings are used to evaluate the approach?",{"text":84,"@type":76},"The approach is evaluated on four ETT-family benchmark datasets (ETTh1, ETTh2, ETTm1, ETTm2) across multiple forecasting settings, including univariate-to-univariate, multivariate-to-univariate, and multivariate-to-multivariate prediction horizons of 96, 192, 336, and 720.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]