[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84317-en":3,"doc-seo-84317-105":29,"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":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},84317,687197207919,"Theodora","https://ap-avatar.wpscdn.com/avatar/a000253d6f5f7c60be?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779446848396160552",8,"Research & Report","Contrastive Order Learning: A General Framework for Ordinal Regression","Contrastive Order Learning (ConOrd) presents a contrastive learning framework tailored for ordinal regression by merging contrastive learning’s batch-wide sample utilization with order learning’s explicit handling of label ordinality. ConOrd introduces a contrastive order loss using soft affinity and disparity weights derived from rank differences, allowing fine-grained modeling of global ordinal relationships across all sample pairs. Experiments on facial age estimation and blind image/video quality assessment show consistent state-of-the-art results and strong generalization across diverse ordinal regression settings.","Contrastive Order Learning: A General Framework for Ordinal Regression  \nChaewon Lee 1 BeomJun Shim 1 Kwang Pyo Choi 2 Chang-Su Kim 1  \narXiv :2607 .08 109v 1 [ cs .LG] 9 Jul 2026  \nAbstract  \nWe propose contrastive order learning (ConOrd), a contrastive learning framework for ordinal regression that integrates the strengths of contrastive learning and order learning. While contrastive learning effectively leverages all samples in a batch, it typically ignores the inherent ordering among rank labels. Conversely, order learning explicitly models label ordinality but often relies on local, margin-based comparisons, limiting its ability to capture global ordinal structure.  \nConOrd addresses these limitations by introducing a contrastive order loss with soft affinity and disparity weights based on rank differences, enabling fine-grained modeling of ordinal relationships across all sample pairs within a batch. Extensive experiments on a range of ordinal regression tasks, including facial age estimation, blind image quality assessment, and blind video quality assessment, demonstrate that ConOrd consistently achieves state-of-the-art performance and generalizes well across diverse ordinal regression scenarios. The source code is available at [https://github.com/cwlee00/ConOrd](https://github.com/cwlee00/ConOrd).  \n1. Introduction  \nOrdinal regression is a task to estimate the discrete or continuous rank of an object instance. For example, facial age estimation aims to predict a person’s age given their facial photograph, while image quality assessment predicts the quality score for an image. It is a fundamental problem frequently arising in many real-world scenarios, including facial age estimation (Moschoglou et al., 2017), health status scoring (Engemann et al., 2022), image and video quality assessment (Ying et al., 2021 ; Hosu et al., 2017), and gaze direction estimation (Wang et al., 2022) .  \nDespite its wide applicability, ordinal regression poses in- 1 School of Electrical Engineering, Korea University, Seoul, Korea 2 Samsung Electronics, Seoul, Korea. Correspondence to:  \nChang-Su Kim \u003C[changsukim@korea.ac.kr](changsukim@korea.ac.kr) >.  \nProceedings of the 43 rd International Conference on Machine Learning, Seoul, South Korea. PMLR 306, 2026 . Copyright 2026 by the author(s) .  \nherent challenges: there is no clear distinction between adjacent ranks, and the semantic gap between neighboring labels can be subtle or ambiguous. It is hence difficult fora machine to learn discriminative representations that accurately reflect the underlying ordinal structure. To address these challenges, various methods have been proposed (Li & Lin, 2007 ; Rothe et al., 2018 ; Geng et al., 2013 ; Diaz & Marathe, 2019) . Recently, order learning techniques (Lim et al., 2020 ; Lee & Kim, 2021 ; Lee et al., 2022) have achieved notable success. Among them, geometric order learning (GOL) (Lee et al., 2022) enforces metric and order constraints to arrange instances according to their ranks inan embedding space. However, as a margin-based pairwise approach, GOL generates no gradient once the margin is satisfied and cannot fully exploit the richer ordinal context available at the batch level, limiting its ability to capture fine-grained ordinal relationships.  \nMeanwhile, supervised contrastive learning (Khosla et al., 2020) has also been extended to ordinal regression, most notably in the RnC algorithm (Zha et al., 2023) . While supervised contrastive learning relies on categorical labels to define positive and negative pairs, RnC constructs rankaware pairs by comparing relative rank differences within a batch. Specifically, RnC selects an anchor and a positive sample, and treats samples whose rank differences from the anchor exceed that of the positive as negatives. However, this hard thresholding collapses ordinal distances into binary decisions, treating samples with both moderately and substantially larger rank gaps equally as negatives. As a res","cbCaiknGlVCsFbhC","https://ap.wps.com/l/cbCaiknGlVCsFbhC","pdf",8083476,1,31,"English","en",105,"# Introduction\n## Contrastive and order learning background\n## Limitations of existing approaches\n## Proposed method: Contrastive Order Learning (ConOrd)\n## Illustration and comparison (Figure 1)","[{\"question\":\"What problem does contrastive order learning (ConOrd) address in ordinal regression?\",\"answer\":\"ConOrd targets the gap between contrastive learning that ignores label ordering and order learning methods that often use local, margin-based comparisons. It aims to capture global ordinal structure more effectively within a batch.\"},{\"question\":\"How does ConOrd define positive and negative relationships between samples?\",\"answer\":\"ConOrd uses a contrastive order loss that assigns soft affinity and disparity weights based on rank differences. This contrasts with methods that use hard thresholds or categorical label treatments.\"},{\"question\":\"Which tasks are used to evaluate ConOrd and what is the outcome?\",\"answer\":\"Experiments cover facial age estimation, blind image quality assessment, and blind video quality assessment. Results show consistently state-of-the-art performance and good generalization across ordinal regression scenarios.\"}]",1784194790,78,{"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":86,"head_meta":88,"extra_data":90,"updated_unix":27},"contrastive-order-learning-a-general-framework-for-ordinal-regression","",{"@graph":35,"@context":85},[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/contrastive-order-learning-a-general-framework-for-ordinal-regression/84317/",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",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does contrastive order learning (ConOrd) address in ordinal regression?","Question",{"text":75,"@type":76},"ConOrd targets the gap between contrastive learning that ignores label ordering and order learning methods that often use local, margin-based comparisons. It aims to capture global ordinal structure more effectively within a batch.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does ConOrd define positive and negative relationships between samples?",{"text":80,"@type":76},"ConOrd uses a contrastive order loss that assigns soft affinity and disparity weights based on rank differences. This contrasts with methods that use hard thresholds or categorical label treatments.",{"name":82,"@type":73,"acceptedAnswer":83},"Which tasks are used to evaluate ConOrd and what is the outcome?",{"text":84,"@type":76},"Experiments cover facial age estimation, blind image quality assessment, and blind video quality assessment. Results show consistently state-of-the-art performance and good generalization across ordinal regression scenarios.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]