[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-663404-105":59,"doc-detail-663404-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","hierarchical-multi-positive-contrastive-learning-for-patent-image-retrieval","Hierarchical Multi-Positive Contrastive Learning for Patent Image Retrieval","","Patent images are technical drawings that represent innovations, and patent image retrieval systems must search huge collections for visually relevant results. Existing approaches still struggle with technical detail and complex semantics and often fail to use patents’ hierarchical relations from the Locarno International Classification (LIC). This work proposes a hierarchical multi-positive contrastive loss that assigns multiple positive pairs per image using LIC taxonomy levels with hierarchy-aware similarity scores. Experiments on DeepPatent2 show improved retrieval across vision and multimodal models, including low-parameter settings.",{"@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/hierarchical-multi-positive-contrastive-learning-for-patent-image-retrieval/663404/",{"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/hierarchical-multi-positive-contrastive-learning-for-patent-image-retrieval/663404.png","ImageObject",300,407,{"name":92,"@type":93},"RuangKosong","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-29","2026-10-28",true,{"@type":102,"interactionType":103,"userInteractionCount":14},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does the paper address in patent image retrieval?","Question",{"text":112,"@type":113},"It addresses the difficulty of retrieving relevant patent images while noting that many methods ignore hierarchical relations defined by the Locarno International Classification (LIC).","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does the proposed hierarchical multi-positive contrastive loss work?",{"text":117,"@type":113},"It forms multiple positive pairs for each patent image in a batch, using hierarchy levels from the LIC to assign varying similarity scores to reflect taxonomic relations.",{"name":119,"@type":110,"acceptedAnswer":120},"What experimental results does the paper report?",{"text":121,"@type":113},"Experiments on the DeepPatent2 dataset with multiple vision and multimodal models show improved retrieval performance, and the method remains effective for low-parameter models with limited hardware resources.","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},663404,1793227289,{"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":14,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":44,"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":129,"read_time":143},962090883219,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","Hierarchical Multi-Positive Contrastive Learning for Patent Image Retrieval  \nKshitij Kavimandan 1, * , Angelos Nalmpantis2, Emma Beauxis-Aussalet1 and Robert-Jan Sips2  \n1 Vrije Universiteit Amsterdam, Amsterdam, The Netherlands 2 TKHAI, Amsterdam, The Netherlands  \nAbstract  \nPatent images are technical drawings that convey information about a patent’s innovation. Patent image retrieval systems aim to search in vast collections and retrieve the most relevant images. Despite recent advances in information retrieval, patent images still pose significant challenges due to their technical intricacies and complex semantic information, requiring efficient fine-tuning for domain adaptation. Current methods neglect patents’hierarchical relationships, such as those defined by the Locarno International Classification (LIC) system, which groups broad categories (e.g.,“furnishing”) into subclasses (e.g.,“seats” and “beds”) and further into specific patent designs. In this work, we introduce a hierarchical multi-positive contrastive loss that leverages the LIC’s taxonomy to induce such relations in the retrieval process. Our approach assigns multiple positive pairs to each patent image within a batch, with varying similarity scores based on the hierarchical taxonomy. Our experimental analysis with various vision and multimodal models on the DeepPatent2 dataset shows that the proposed method enhances the retrieval results. Notably, our method is effective with low-parameter models, which require fewer computational resources and can be deployed on environments with limited hardware.  \nKeywords  \nInformation Retrieval, Patent Image Retrieval, Hierarchical Multi-Positive Contrastive Learning  \n1. Introduction  \nPatent images are technical drawings that illustrate the novelty of a patent, often conveying their details more effectively than natural language written in text [1] . Thereby, technical patent reports are typically accompanied by multiple images capturing different aspects of the invention. With the rapidly growing volume of patents, efficient patent image retrieval systems are becoming an essential component for searching these vast collections.  \nMany advances in information retrieval have been largely driven by the power of attention based models [2, 3] and the knowledge acquired during extensive pretraining phases, mainly focused on the language domain. While similar models, such as Vision Transformer (ViT) [4] and ResNet [5], have provided remarkable results on a plethora of vision tasks, they still fall short when processing technical drawings since their pretraining mainly involves natural images [6] . In response, to address this domain shift, researchers have released specialized sketch datasets [7, 8] that facilitate model fine-tuning on such images. Similarly, large scale datasets containing patent images have emerged to address their unique intricacies and enable the development of efficient patent image retrieval methods.  \nDeepPatent [1] was the first large scale dataset designed for training and evaluating patent image retrieval systems, comprising over 350, 000 images across 45, 000 patents, enabling the development of PatentNet, which exhibited significant improvements in patent image retrieval. Additionally, several studies investigated the generation of synthetic text descriptions by leveraging the zero-shot capabilities of (vision) language models [9, 10], allowing the application of multimodal models, such as CLIP [11], on patent image retrieval. Inspired by DeepPatent, DeepPatent2 [12] provided an extension of  \n6th Workshop on Patent Text Mining and Semantic Technologies (PatentSemTech) 2025  \n* Corresponding author.  \n$ [kshitijk3188@gmail.com](kshitijk3188@gmail.com) (K. Kavimandan); [a.nalmpantis@tkh.ai](a.nalmpantis@tkh.ai) (A. Nalmpantis); [e.m.a.l.beauxisaussalet@vu.nl](e.m.a.l.beauxisaussalet@vu.nl)  \n(E. Beauxis-Aussalet); [r.sips@tkh.ai](r.sips@tkh.ai) (R. Sips)  \n􀀚 0009-0002-3760-5882 (K. Kavimandan","cbCaincqpFYfAxJe","https://ap.wps.com/l/cbCaincqpFYfAxJe","pdf",2695680,"English","# Abstract\n# Keywords\n# 1. Introduction\n## Patent image retrieval challenges\n## Domain shift and specialized datasets\n## DeepPatent and DeepPatent2 background\n## Locarno International Classification and hierarchy\n## Proposed hierarchical multi-positive contrastive learning","[{\"question\":\"What problem does the paper address in patent image retrieval?\",\"answer\":\"It addresses the difficulty of retrieving relevant patent images while noting that many methods ignore hierarchical relations defined by the Locarno International Classification (LIC).\"},{\"question\":\"How does the proposed hierarchical multi-positive contrastive loss work?\",\"answer\":\"It forms multiple positive pairs for each patent image in a batch, using hierarchy levels from the LIC to assign varying similarity scores to reflect taxonomic relations.\"},{\"question\":\"What experimental results does the paper report?\",\"answer\":\"Experiments on the DeepPatent2 dataset with multiple vision and multimodal models show improved retrieval performance, and the method remains effective for low-parameter models with limited hardware resources.\"}]","Hierarchical Multi-Positive Contrastive Learning for Patent Image Retrieval | PDF",23]