[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-514903-105":59,"doc-detail-514903-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","attention-is-not-everything-efficient-alternatives-for-vision-review-of-non-transformer-methods-in-computer-vision","Attention Is not Everything - Efficient Alternatives for Vision - Review of Non-Transformer Methods in Computer Vision","","Non-Transformer approaches remain highly relevant as efficient competitors to Transformer-based computer vision models. This review surveys methods developed as alternatives, focusing on convolution-based, MLP-based, state-space and structured models, graph-based methods, implicit neural representations, energy- and kernel-driven techniques, probabilistic frameworks, and classical CV–deep hybrids. Forty papers are selected to map the taxonomy and evaluate efficiency, scaling behavior, interpretability, and robustness, outlining challenges and opportunities for future research.",{"@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":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/attention-is-not-everything-efficient-alternatives-for-vision-review-of-non-transformer-methods-in-computer-vision/514903/",{"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/attention-is-not-everything-efficient-alternatives-for-vision-review-of-non-transformer-methods-in-computer-vision/514903.png","ImageObject",300,407,{"name":92,"@type":93},"WPS_1790064749","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-16","2026-10-06",true,{"@type":102,"interactionType":103,"userInteractionCount":19},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"Why does the paper focus on non-Transformer computer vision methods?","Question",{"text":112,"@type":113},"Transformers have major limitations such as high data requirements and heavy computation. The paper argues that non-Transformer models can mitigate these issues while staying competitive.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"What taxonomy categories does the review organize?",{"text":117,"@type":113},"The review groups approaches into categories including convolution-based, MLP-based, graph-based, state-space and structured models, implicit neural representations, energy-based and kernel-driven methods, probabilistic frameworks, and classical CV–deep hybrids.",{"name":119,"@type":110,"acceptedAnswer":120},"How many papers are selected, and what is the selection intent?",{"text":121,"@type":113},"A total of 40 papers are chosen to cover recent progress (from 2023 to 2025) and ensure methodological importance, relevance, and citation diversity across the identified taxonomies.","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},514903,1792231900,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":19,"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,"preview_page_count":19,"preview_urls":146},3985747859154,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","Attention Is not Everything: Efficient Alternatives for Vision  \narXiv :2604 . 17439v1 [ cs .CV] 19 Apr 2026  \nNur Mohammad Kazi Ahsanullah University of Science and Technology Dhaka, Bangladesh  \n[nurmohammadkazi2002@gmail.com](nurmohammadkazi2002@gmail.com)  \nIbteshum Khaled Ahsanullah University of Science and Technology Dhaka, Bangladesh  \n[ibteshum.khaled123@gmail.com](ibteshum.khaled123@gmail.com)  \nMd. Luthful Hasan Galib Ahsanullah University of Science and Technology Dhaka, Bangladesh  \n[omi673galib@gmail.com](omi673galib@gmail.com)  \nAli Faruk Shihab Ahsanullah University of Science and Technology Dhaka, Bangladesh  \n[faruk.aka.shihab@gmail.com](faruk.aka.shihab@gmail.com)  \nAbstract  \nRecently computer vision has seen advancements mainly thanks to Transformer-based models. However many nonTransformer methods are still doing well being a direct competition of Transformer-based models. This review tries to present a comprehensive taxonomy of such methods and organize these methods into categories like convolutionbased models, MLP-based models, state-space-based and more. These methods are looked at in terms of how efficient they are, how well they scale, how easy they are to understand and how robust they are. A total of 40 papers were chosen for this study. The goal is to give a view of non-Transformer methods and find out what challenges and opportunities exist for future computer vision research.  \nKeywords: Computer Vision, Image Segmentation, Convolutional Neural Networks, State Space Models, Uncertainty-Aware Methods  \n1. Introduction  \nComputer vision is an ever changing field with different ways of thinking and development of new models. At first people used a lot of manual help and algorithms and then deep learning came along. Convolutional neural networks became popular because they had better understanding of relationships in a space than previous methods and relied less on manual input. Recently people have been using models that are good at understanding how things change  \nMd. Rakibul Islam Ahsanullah University of Science and Technology  \nDhaka, Bangladesh  \n[rakib.cse@aust.edu](rakib.cse@aust.edu)  \nover time and models that can handle continuous signals, like sound or video. They are also using methods to optimize things and frameworks that use probability to estimate how sure we are about something. People are also combining computer vision methods with deep learning to make models that are easier to understand and can work better indifferent situations.  \nThe introduction of Transformer architectures has significantly influenced modern machine learning fields, specifically computer vision. By enabling global context modeling through self-attention mechanisms, Transformers achieved impressive results that reshaped the research field. Starting with the work “Attention Is All You Need” [1], Transformers were adapted to vision tasks through models such as the Vision Transformer (ViT) [2], which demonstrated that pure attention-based architectures can achieve competitive performance when trained on large-scale datasets. Following works, including Data-efficient Image Transformers (DeiT) [3] and the Swin Transformer [4], further improved performance establishing Transformers as a dominant paradigm in the field. However methods that use Transformers have some problems. They need a lot of data to work well. They use a lot of computational power and memory. They do not have built-in functionality that helps when there is not much data available. Non-Transformer models complements them as they are capable of mitigating stated limitations. Non-transformer approaches remain highly relevant and continue to evolve as competitive alternatives. It can be highly practical when transformer based models are not compatible with the given task environ-  \nment. This paper presents a comprehensive taxonomy of such methods, including convolution-based models, MLPbased models, graph-based models, state-space and structured ","cbCaiatoCcVV9O0J","https://ap.wps.com/l/cbCaiatoCcVV9O0J","pdf",289457,21,"English","# 1. Introduction\n# 2. Convolution-Based Approach\n## 2.1. State-of-the-Art Convolution-Based Methods","[{\"question\":\"Why does the paper focus on non-Transformer computer vision methods?\",\"answer\":\"Transformers have major limitations such as high data requirements and heavy computation. The paper argues that non-Transformer models can mitigate these issues while staying competitive.\"},{\"question\":\"What taxonomy categories does the review organize?\",\"answer\":\"The review groups approaches into categories including convolution-based, MLP-based, graph-based, state-space and structured models, implicit neural representations, energy-based and kernel-driven methods, probabilistic frameworks, and classical CV–deep hybrids.\"},{\"question\":\"How many papers are selected, and what is the selection intent?\",\"answer\":\"A total of 40 papers are chosen to cover recent progress (from 2023 to 2025) and ensure methodological importance, relevance, and citation diversity across the identified taxonomies.\"}]","Attention Is not Everything - Efficient Alternatives for Vision - Review of Non-Transformer Methods in Computer Vision | PDF",1791303207,53,[147,151,155,159],{"page_no":8,"url":148,"width":149,"height":150},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/14ff8e60cc8a5e483467c0b81229b41d.webp",983,1272,{"page_no":14,"url":152,"width":153,"height":154},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/1bbea100cfa76464901f653b56dc0ae3.webp",982,1270,{"page_no":81,"url":156,"width":157,"height":158},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/33c053c11ee72efcdb77c9fd5aed0f64.webp",981,1269,{"page_no":19,"url":160,"width":161,"height":162},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/8be0b34242638b78b98eb1314c64a6bc.webp",980,1268]