[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82364-en":3,"doc-seo-82364-105":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},82364,687197207919,"Theodora","https://ap-avatar.wpscdn.com/avatar/a000253d6f5f7c60be?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779446848396160552",8,"Research & Report","Foveation-Guided Dynamic Token Selection for Robust and Efficient Vision Transformers","The human visual system combines foveated sampling and eye movements to perceive scenes efficiently while conserving metabolic energy and computational resources. Inspired by this robustness and adaptability, the Foveated Dynamic Transformer (FDT) integrates foveation and dynamic fixation into a vision transformer for adaptive token selection. FDT improves resilience to noise and adversarial attacks without dedicated training by filtering irrelevant information via fixation and encoding multi-scale context through foveated embeddings. With a 50% fixation-budget, FDT reaches higher accuracy than DeiT-S (81.9% vs. 80.9%) while cutting multiply-accumulate operations by 34.57%.","arXiv :2607 .09480v1 [ cs .CV] 10 Jul 2026  \nHighlights  \nFoveation-Guided Dynamic Token Selection for Robust and Efficient Vision Transformers  \nIbrahim Batuhan Akkaya, Kishaan Jeeveswaran, Bahram Zonooz, Elahe Arani  \n• FDT integrates foveation and fixation into vision transformers.  \n• Dynamic fixation selects informative tokens in a single feedforward pass.  \n• Foveated tokens encode multi-scale context for adaptive attention.  \n• FDT improves robustness without adversarial or corruption training.  \n• At 50% fixation budget, FDT reduces MACs by 34 .57% .  \nFoveation-Guided Dynamic Token Selection for Robust and Efficient Vision Transformers  \nIbrahim Batuhan Akkayaa,∗, Kishaan Jeeveswaranb , Bahram Zonoozb,1 ,  \nElahe Aranib,1  \na Advanced Research Lab, NavInfo Europe, Eindhoven, 5657 DB, Netherlands b Department of Mathematics and Computer Science, Eindhoven University of  \nTechnology, Eindhoven, 5612 AZ, Netherlands  \nAbstract  \nThe human visual system (HVS) employs foveated sampling and eye movements to achieve efficient perception, conserving both metabolic energy and computational resources. Drawing inspiration from this robustness and adaptability, we introduce the Foveated Dynamic Transformer (FDT), a foveation-guided dynamic token-selection architecture that integrates these mechanisms into a vision transformer framework. The FDT exhibits strong resilience to various types of noise and adversarial attacks, despite not being explicitly trained for such challenges. This inherent robustness is achieved through the use of fixation and foveation modules: the fixation module identifies fixation points to filter out irrelevant information, while the foveation module generates foveated embeddings with multi-scale information. At the 50% fixation-budget setting, FDT achieves higher accuracy than DeiT-S (81.9% vs. 80.9%) while reducing multiply-accumulate operations by 34.57%,  \n∗ Corresponding author  \nEmail address: [bthakkaya@gmail.com](bthakkaya@gmail.com) (Ibrahim Batuhan Akkaya)  \n1 Equal contribution  \nhighlighting one operating point on its accuracy-efficiency trade-off. These attributes position FDT as an HVS-inspired step toward artificial neural networks that combine adaptive computation with improved resilience.1 Keywords: Vision Transformer, Foveated Vision, Dynamic Token Selection, Adversarial Robustness, Human Visual System, Efficient Inference  \n1. Introduction  \nRecent studies indicate that deep neural networks and the human brain interpret the environment differently, with the human visual system (HVS) dynamically filtering task-irrelevant information to focus on potential objects of interest, a selective mechanism that contributes to perceptual stability and resilience against noisy or misleading inputs [1–4] . The retina contains photoreceptors, with the fovea–a high spatial resolution area–playing a key role in color perception and visual detail recognition [5] . The highest photoreceptor density at the fovea decreases with eccentricity, resulting in a variable-resolution image transmitted to the brain, a phenomenon known as foveation, highlighting HVS’s multi-resolution perception. Such mechanisms suggest that vision models could benefit from integrating spatially adaptive processing strategies inspired by the HVS. In addition to efficiency and robustness, such biologically inspired mechanisms enhance spatial awareness, allowing models to preserve fine details in salient regions while maintaining global scene coherence, similar to how the human visual system balances local and global perception. Studying HVS to enhance deep neural network design is therefore a promising research avenue for developing intelligent agents.  \n1The code will be shared upon acceptance.  \nFigure 1: Illustration comparing the fixations of the human visual system (left; Yarbus et al. [11]) and an ANN (right; our method) . Areas outside the fixations are blurred to highlight regions of interest. The sequence of eye movements is i","cbCaigYRVgVKGQ2p","https://ap.wps.com/l/cbCaigYRVgVKGQ2p","pdf",9065913,1,51,"English","en",105,"# Highlights\n# Abstract\n# Keywords\n# Introduction","[{\"question\":\"What are the reported accuracy and compute trade-offs for FDT?\",\"answer\":\"At a 50% fixation budget, FDT improves accuracy over DeiT-S (81.9% vs. 80.9%) while reducing multiply-accumulate operations by 34.57%, showing one operating point on its accuracy-efficiency trade-off.\"}]",1784179929,129,{"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},"foveation-guided-dynamic-token-selection-for-robust-and-efficient-vision-transformers","",{"@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/foveation-guided-dynamic-token-selection-for-robust-and-efficient-vision-transformers/82364/",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 are the reported accuracy and compute trade-offs for FDT?","Question",{"text":75,"@type":76},"At a 50% fixation budget, FDT improves accuracy over DeiT-S (81.9% vs. 80.9%) while reducing multiply-accumulate operations by 34.57%, showing one operating point on its accuracy-efficiency trade-off.","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"]