[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-328041-105":59,"doc-detail-328041-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","adastop-cost-aware-early-stopping-for-dnn-test-selection","AdaStop - Cost-Aware Early Stopping for DNN Test Selection","","Existing methods for testing deep neural networks mostly allocate a fixed labeling budget to inputs likely to expose faults, but real-world budgets are hard to choose because too little testing misses failures and too much wastes labeling cost. AdaStop models DNN testing as a cost–benefit decision: labeling an input costs c and discovering a fault yields value v. It estimates the marginal fault discovery rate during testing and stops when the estimated rate drops below τ=c/v. Experiments show 65–84% recall with only 9–31% of the budget.",{"@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/adastop-cost-aware-early-stopping-for-dnn-test-selection/328041/",{"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/adastop-cost-aware-early-stopping-for-dnn-test-selection/328041.png","ImageObject",300,407,{"name":92,"@type":93},"Eden","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-31","2026-09-21",true,{"@type":102,"interactionType":103,"userInteractionCount":24},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does AdaStop address in DNN test selection?","Question",{"text":112,"@type":113},"AdaStop addresses the stopping problem in DNN testing: deciding when to stop labeling test inputs so that fault discovery remains efficient under labeling costs.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does AdaStop determine when to stop labeling?",{"text":117,"@type":113},"AdaStop estimates the marginal fault discovery rate during testing and stops when the estimated rate falls below the cost-adjusted threshold τ=c/v.",{"name":119,"@type":110,"acceptedAnswer":120},"What do the experiments show about labeling budget efficiency?",{"text":121,"@type":113},"Across multiple datasets, architectures, and selection strategies, AdaStop discovers 65–84% of faults using only 9–31% of the labeling budget.","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},328041,1790517238,{"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":24,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":29,"language":139,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":67,"update_tm":143,"read_time":144,"preview_page_count":4},1374391974468,"https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0","AdaStop: Cost-Aware Early Stopping for DNN Test Selection  \nBonan Shen 1 , Wei-Jung Huang 1 , Xin Liu 1 , Jiazhou Gao 1 and Tao Ning 1  \narXiv :2607 .0546 1v 1 [ cs .LG] 6 Jul 2026  \nAbstract—Existing methods for testing deep neural networks (DNNs) primarily prioritize test inputs likely to reveal model faults under a fixed labeling budget. In practice, choosing that budget is difficult: too little testing misses failures, while too much incurs unnecessary labeling costs. This work studies the stopping problem in DNN testing. We formulate testing as a cost–benefit decision process in which labeling an input incurs cost c and discovering a fault yields value v. Based on this formulation, we introduce AdaStop, a framework that estimates the marginal fault discovery rate during testing and stops labeling when the estimated rate falls below the threshold τ = c/v. Experiments across multiple datasets, architectures, and selection strategies show that 65–84% of faults can be discovered using only 9–31% of the labeling budget.  \nI. INTRODUCTION  \nDeep neural networks (DNNs) are increasingly deployed in domains where reliability is critical, including autonomous vehicles [24], medical diagnosis [25], and financial systems [26] . Ensuring reliability requires systematic testing to identify test inputs on which the model makes incorrect predictions—commonly referred to as faults. In many settings, however, verifying predictions requires ground-truth labels obtained through costly human annotation by domain experts [27] .  \nThis creates a fundamental and practical challenge: when should we stop labeling test inputs? Stopping too early risks missing faults that could lead to system failures; stopping too late wastes labeling resources on inputs that are unlikely to reveal new faults.  \nExisting test selection methods primarily focus on the selection problem—determining which inputs to label first—while the complementary stopping problem has received comparatively little attention. We identify three key limitations:  \n1) Lack of principled stopping criterion. Methods like DeepGini [9], TestRank [10], ATS [11], and DeepSample [12] typically evaluate performance under predetermined labeling budgets without providing guidance on when testing should terminate.  \n2) Arbitrary budget selection. In the absence of stopping criteria, practitioners must choose labeling budgets heuristically. However, appropriate budgets depend on factors such as model quality and dataset characteristics, which are often unknown beforehand.  \n3) Unaddressed cost–benefit trade-off. Labeling each input incurs cost, while discovering a fault provides value. Existing approaches focus on optimizing the  \n1Independent Researcher {shenbonan2, william.wj.huang, iamxinliu, gjz140103, [ntgd1102](ntgd1102}@gmail.com)[}](ntgd1102}@gmail.com)[@gmail.com](ntgd1102}@gmail.com)  \norder of test selection, such as prioritizing inputs that are more likely to reveal model errors, but do not explicitly consider this cost–benefit trade-off [9] .  \nAlthough stopping criteria exist in related domains, such as the SAFE procedure [18] for document screening and convergence-based stopping [13] for active learning, these methods have not been adapted to DNN test selection, where the goal is efficient fault discovery.  \nA key empirical observation motivates our approach: test selection often exhibits diminishing returns. Inputs with higher uncertainty, which are typically selected first, tend to reveal faults at higher rates. As testing progresses, the remaining inputs yield progressively fewer faults per labeled instance.  \nThis behavior suggests that the value of additional labeling decreases over time, creating a natural point at which further testing becomes inefficient. Based on this observation, we propose AdaStop, a cost-aware framework that terminates testing when the marginal fault discovery rate p (t) falls below a cost-adjusted threshold τ = c/v, where c denotes the labeling cost and ","cbCaik6CImRbK13d","https://ap.wps.com/l/cbCaik6CImRbK13d","pdf",289983,"English","# Introduction\n## Problem challenge: when to stop labeling\n## Limitations of existing test selection methods\n# Problem Formulation\n## Setting and objective\n## Optimal stopping threshold\n# AdaStop Framework\n## Components and cost-aware stopping rule\n# Contributions","[{\"question\":\"What problem does AdaStop address in DNN test selection?\",\"answer\":\"AdaStop addresses the stopping problem in DNN testing: deciding when to stop labeling test inputs so that fault discovery remains efficient under labeling costs.\"},{\"question\":\"How does AdaStop determine when to stop labeling?\",\"answer\":\"AdaStop estimates the marginal fault discovery rate during testing and stops when the estimated rate falls below the cost-adjusted threshold τ=c/v.\"},{\"question\":\"What do the experiments show about labeling budget efficiency?\",\"answer\":\"Across multiple datasets, architectures, and selection strategies, AdaStop discovers 65–84% of faults using only 9–31% of the labeling budget.\"}]","AdaStop - Cost-Aware Early Stopping for DNN Test Selection | PDF",1789978568,15]