[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82194-en":3,"doc-seo-82194-105":28,"detail-sidebar-cat-0-en-105":90},{"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":11,"language":21,"language_code":22,"site_id":23,"html_lang":22,"table_of_contents":24,"faqs":25,"seo_title":13,"seo_description":14,"update_tm":26,"read_time":27},82194,2336464648746,"Skyler","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","A Coreset Selection Framework with Ensemble Aggregation for Image Classification","Rapid growth of image datasets increases the time and memory costs of training while making it difficult to choose representative subsets, since individual sample contributions are unclear and model behavior can vary across datasets and runs. The framework combines coreset selection with ensemble aggregation over independently sampled training subsets. SCOre-Stratified Selection (SCOSS) partitions data by score intervals and samples from each, while an ensemble aggregates multi-run predictions. Experiments on SGC and SVM show efficient accuracy trade-offs and strong performance with fewer labeled samples, with public code released.","A Coreset Selection Framework with Ensemble Aggregation for Image Classification  \nPedro Rocha Dantas 1 , Lucas Pascotti Valem 1  \n1 Institute of Mathematics an˜d Computer Science (ICMC)  \nUnive˜ rsity of Sao Paulo (USP)  \nSao Carlos – SP – Brazil  \n[pedrord@usp.br](pedrord@usp.br) , [lucas@icmc.usp.br](lucas@icmc.usp.br)  \nPreprint. Accepted at the Workshop de Trabalhos de Alunos de Graduac¸ o (WTAG) of the Brazilian Symposium on Databases (SBBD 2026) .  \narXiv :2607 .09100v1 [ cs .CV] 10 Jul 2026  \nAbstract. The rapid growth of image data has produced large-scale datasets, raising concerns about the time and memory costs of model training. Selecting representative training subsets, however, remains challenging: individual sample contributions are unclear, and model behavior varies across datasets and runs. We address these challenges with a framework that combines core set selection with an ensemble aggregation over multiple runs. For coreset selection, we propose SCOre-Stratified Selection (SCOSS), which partitions the training data into intervals based on a chosen score and samples from each in terval. The ensemble combines predictions from multiple runs, each performed on an independently sampled training subset. As baselines, we use moderate and random selection, each in original and class-balanced versions. We assess the framework with Simple Graph Convolution (SGC) and Support Vector Machine (SVM) classifiers under different sampling ratios. Experiments show that SCOSS is competitive with baselines, often the best choice for SGC, and enables favorable trade-offs between accuracy and efficiency. On the finegrained dataset, SGC with SCOSS outperforms SVMs when using fewer la beled samples. The code and supplementary materials are publicly available at [scoss.lucasvalem.com](scoss.lucasvalem.com).  \n1. Introduction  \nThe growing availability of large-scale datasets has driven progress in image classification, but has also raised concerns about the computational cost of model training in terms of memory and processing time [Moser et al. 2026] . This concern is even more pressing in transductive models such as Graph Convolutional Networks (GCNs) [Gkarmpounis et al. 2024], where the entire graph structure participates in training and inference. Therefore, reducing the amount of training data while preserving classification performance has become a central challenge in the field [Guo et al. 2022] .  \nA common strategy to address this problem is coreset selection, which aims to identify representative subsets of the original dataset to reduce training cost [Feldman 2019, Guo et al. 2022] . Despite its potential, selecting informative samples remains difficult, as the contribution of each instance to model performance isnot always clear, especially in graph-structured data where samples are interdependent [Wu et al. 2019] . Moreover, model behavior may vary across datasets and training runs, making the selection process inherently unstable. These challenges motivate the  \nsearch for approaches capable of producing representative subsets while improving robustness and consistency across executions [Moser et al. 2026, Sagi and Rokach 2018] .  \nSeveral approaches have been proposed for coreset selection in classification tasks [Feldman 2019] . Random sampling and Moderate Coreset [Xia et al. 2023] serve as common baselines: the former selects instances uniformly at random, while the latter focuses on samples close to the score median. More structured methods include one inspired by stratified sampling that improves data coverage [Zheng et al. 2023], and another that allocates the budget per class based on within-class difficulty [Tsai et al. 2025] . The method proposed in this work combines both ideas, applying stratified sampling over the full score distribution with class-balanced budget allocation.  \nIn this work, we propose a framework that combines coreset selection with an ensemble strategy to improve robustness and predicti","cbCaicxbTFtNrFzs","https://ap.wps.com/l/cbCaicxbTFtNrFzs","pdf",1030364,1,"English","en",105,"# Introduction\n# Methodology\n## Framework Overview","[{\"question\":\"Why is coreset selection difficult for image classification on large-scale datasets?\",\"answer\":\"Individual samples’ contributions are not always clear, and model behavior can vary across datasets and training runs, making selection unstable.\"},{\"question\":\"What is SCOSS (SCORE-Stratified Selection) in the proposed framework?\",\"answer\":\"SCOSS partitions training data into score-based intervals and samples instances from each interval to build a representative coreset.\"},{\"question\":\"How does the ensemble aggregation improve robustness in this framework?\",\"answer\":\"Multiple runs are trained on independently sampled training subsets, and their predictions are aggregated to reduce variability across executions.\"}]",1784178730,20,{"code":4,"msg":29,"data":30},"ok",{"site_id":23,"language":22,"slug":31,"title":13,"keywords":32,"description":14,"schema_data":33,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":26},"a-coreset-selection-framework-with-ensemble-aggregation-for-image-classification","",{"@graph":34,"@context":84},[35,52,67],{"@type":36,"itemListElement":37},"BreadcrumbList",[38,42,46,49],{"item":39,"name":40,"@type":41,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":43,"name":44,"@type":41,"position":45},"https://docshare.wps.com/document/","Document",2,{"item":47,"name":12,"@type":41,"position":48},"https://docshare.wps.com/document/research-report/",3,{"item":50,"name":13,"@type":41,"position":51},"https://docshare.wps.com/document/a-coreset-selection-framework-with-ensemble-aggregation-for-image-classification/82194/",4,{"url":50,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":22,"description":14,"dateModified":60,"datePublished":61,"encodingFormat":59,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":39,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":20},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"Why is coreset selection difficult for image classification on large-scale datasets?","Question",{"text":74,"@type":75},"Individual samples’ contributions are not always clear, and model behavior can vary across datasets and training runs, making selection unstable.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"What is SCOSS (SCORE-Stratified Selection) in the proposed framework?",{"text":79,"@type":75},"SCOSS partitions training data into score-based intervals and samples instances from each interval to build a representative coreset.",{"name":81,"@type":72,"acceptedAnswer":82},"How does the ensemble aggregation improve robustness in this framework?",{"text":83,"@type":75},"Multiple runs are trained on independently sampled training subsets, and their predictions are aggregated to reduce variability across executions.","https://schema.org",{"og:url":50,"og:type":86,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":88,"canonical":50},"index,follow",{"doc_id":7,"site_id":23},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,126,129,133],{"id":20,"doc_module":4,"doc_module_name":44,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":45,"doc_module":4,"doc_module_name":44,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":51,"doc_module":4,"doc_module_name":44,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":44,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":44,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":44,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":44,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":44,"category_name":124,"show_sort_weight":27,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":27,"doc_module":4,"doc_module_name":44,"category_name":127,"show_sort_weight":27,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":44,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":44,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]