[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83584-en":3,"doc-seo-83584-105":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},83584,8796095360427,"Lucas Martin","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","Benchmarking Federated Learning and Knowledge Distillation for Point Cloud Classification","Privacy-sensitive, resource-constrained 3D point cloud classification faces two coupled constraints: training data cannot be centralized, and the final model must run efficiently on limited edge hardware. This work introduces a multi-seed benchmark that jointly evaluates federated learning (FL) and knowledge distillation (KD) across 13 FL algorithms and 10 KD objectives, covering all teacher–objective combinations per dataset and repeating each setting over three random seeds. Results analyze federated degradation and an evaluation pitfall on ModelNet40 and a clinical craniosynostosis dataset, then recommend label-free distillation to ensure accuracy reflects the federated teacher rather than proxy labels.","Benchmarking Federated Learning & Knowledge  \narXiv :2607 .01272v1 [ cs .GR] 30 Jun 2026  \nDistillation for Point Cloud Classification  \nAizierjiang Aiersilan  \nUniversity of Macau  \n[ezharjan@outlook.com](ezharjan@outlook.com)  \nAbstract. Deploying 3D point cloud analysis in privacy-sensitive and resource-constrained settings faces two coupled barriers: data cannot be centralized for training, and the trained model must run on limited edge hardware. We present a multi-seed benchmark that jointly evaluates federated learning (FL) and knowledge distillation (KD) for 3D point cloud classification. It spans thirteen FL algorithms and ten KD objectives, supporting their full 130-pair teacher–objective cross-product per dataset; every standardized configuration is repeated over three random seeds for 504 training runs in total, with the complete combined grid evaluated at multi-seed scale on the clinical dataset. We characterize federated degradation and the combined-pipeline pitfall on ModelNet40, then validate them on a real-world clinical craniosynostosis dataset of patient head shapes, where the privacy and edge-deployment stakes are concrete. Wereport three findings. First, under extreme non-independent and identically distributed (non-IID) label skew, standalone FL degrades sharply: on ModelNet40 the strongest method reaches only 76.32% against a 92.26% centralized reference, on the clinical data the best reaches 75.83% against 100%, and the four server-side optimizers collapse to near the chance level; the best algorithm differs by dataset, so none is universally robust. Second, distillation compresses the teacher into a student 74.51% smaller and roughly twice as fast at inference, with five of the seven objectives evaluated on ModelNet40 matching or surpassing the 92.44% teacher. Third, the combined pipeline exposes an evaluation pitfall: when distillation keeps a hard-label cross-entropy term on a labeled proxy split, a collapsed federated teacher at 8.50% paired with Logit-MSE still yields a 92.94% student. This 84.4-point gap reflects the proxy labels rather than the federated model, and the hard-label term reuses the very labels whose privacy motivated federation. Objectives without a hard-label term instead track teacher quality (r ≈ 0.99 on the clinical grid) and collapse when the teacher does. We therefore recommend evaluating FL-KD pipelines with label-free distillation, so that the reported accuracy reflects the federated teacher rather than the proxy.  \nKeywords: Federated learning · Knowledge distillation · 3D point cloud · Model compression · Non-IID · Privacy-preserving learning · Medical imaging · Benchmark  \nBenchmark & Code: [https://ezharjan.github.io/FLKD3DBenchmark](https://ezharjan.github.io/FLKD3DBenchmark)  \n2 Aizierjiang Aiersilan  \n1 Introduction  \n3D point cloud analysis underpins applications including autonomous driving, robotics, augmented reality, and medical imaging, where sensor data are naturally represented as irregular point sets in Euclidean space. Architectures such as PointNet [34] and PointNet++ [35] achieve strong classification performance on standard benchmarks, yet their deployment in privacy-sensitive, resourceconstrained settings faces two challenges that have not been studied together in one benchmark.  \nThe first challenge is data privacy. In medical [25, 39 , 44], industrial [9, 27 , 32], and multi-institutional settings [21, 26 , 53], raw sensor data cannot be centralized because of regulatory or competitive constraints. Federated learning (FL) [30] addresses this by coordinating model training across clients without exchanging raw data, using iterative local updates and global aggregation. Even so, gradient inversion attacks [6, 61] show that transmitted gradients can still leak private data, and the broader open problems in FL are surveyed in [15] . The statistical heterogeneity of data across clients, the non-IID problem [12, 14 , 28], degrades convergence and final accura","cbCaipRLZkMiac74","https://ap.wps.com/l/cbCaipRLZkMiac74","pdf",2083992,4,1,62,"English","en",105,"# Introduction\n## Data privacy challenge\n## Computational efficiency at deployment\n## Motivation for jointly evaluating FL and KD","[{\"question\":\"What does the document contribute to evaluating FL and KD for point cloud classification?\",\"answer\":\"It proposes a multi-seed benchmark that jointly evaluates federated learning and knowledge distillation for 3D point cloud classification across many FL algorithms and KD objectives, repeating standardized configurations over multiple random seeds.\"},{\"question\":\"How does non-IID label skew affect standalone federated learning results?\",\"answer\":\"Under extreme non-IID label skew, standalone FL degrades sharply compared with a centralized reference, and server-side optimizers collapse toward near-chance performance.\"},{\"question\":\"Why can the combined FL-KD pipeline produce misleading student accuracy?\",\"answer\":\"If distillation uses a hard-label cross-entropy term on a labeled proxy split, a collapsed federated teacher can still yield high student accuracy due to proxy labels; the authors recommend label-free distillation to better reflect the federated teacher.\"}]",1784189009,156,{"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},"benchmarking-federated-learning-and-knowledge-distillation-for-point-cloud-classification","",{"@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/benchmarking-federated-learning-and-knowledge-distillation-for-point-cloud-classification/83584/",{"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","2026-07-24","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 does the document contribute to evaluating FL and KD for point cloud classification?","Question",{"text":75,"@type":76},"It proposes a multi-seed benchmark that jointly evaluates federated learning and knowledge distillation for 3D point cloud classification across many FL algorithms and KD objectives, repeating standardized configurations over multiple random seeds.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does non-IID label skew affect standalone federated learning results?",{"text":80,"@type":76},"Under extreme non-IID label skew, standalone FL degrades sharply compared with a centralized reference, and server-side optimizers collapse toward near-chance performance.",{"name":82,"@type":73,"acceptedAnswer":83},"Why can the combined FL-KD pipeline produce misleading student accuracy?",{"text":84,"@type":76},"If distillation uses a hard-label cross-entropy term on a labeled proxy split, a collapsed federated teacher can still yield high student accuracy due to proxy labels; the authors recommend label-free distillation to better reflect the federated teacher.","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"]