[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83889-en":3,"doc-seo-83889-105":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},83889,8796095461564,"Liam","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Beyond Independent Labels: Schwartz-Geometry Decoding for Human Value Detection","Human value detection is commonly implemented as sentence-level multi-label classification across 19 refined Schwartz values, often predicted with independent labels. Schwartz theory instead models values as a circular motivational continuum, where adjacent values are compatible and opposing values are in tension. This work operationalizes the structure as output-space geometry and tests it as a soft bias. A DeBERTa-v3-base setup compares training-time geometry objectives and a post-hoc Schwartz-aware energy decoder. The decoder improves theory-aware label-set coherence without changing Macro-F1 or Micro-F1.","Beyond Independent Labels: Schwartz-Geometry Decoding for  \nHuman Value Detection  \nVíctor Yeste 1 ,2 ,∗ and Paolo Rosso 1 ,3  \n1PRHLT Research Center, Universitat Politècnica de València, Spain  \n2 School of Science, Engineering and Design, Universidad Europea de Valencia, Spain  \n3Valencian Graduate School and Research Network of Artificial Intelligence (ValgrAI)  \n∗ Corresponding author: vicyesmo@upv .es  \narXiv :2607 .05052v 1 [ cs .CL] 6 Jul 2026  \nAbstract  \nHuman value detection is commonly formulated as sentence-level multi-label classification over the 19 refined Schwartz values, typically predicted as independent labels. Schwartz theory, however, describes them as a circular motivational continuum, in which adjacent values are compatible and opposing values are in tension. We ask whether this structure can be operationalized as an explicit output-space geometry and used as a soft bias rather thana hard constraint. On a DeBERTa-v3-base classifier, we compare two ways of injecting it: training-time geometry-aware objectives and a post-hoc Schwartz-aware energy decoder that scores whole label sets jointly. Across five seeds, trainingtime geometry gives only limited gains—no larger for the true continuum than for a random ordering—whereas the decoder makes label sets more coherent with the continuum—on theory-aware coherence metrics we introduce—at no cost to MacroF1 or Micro-F1 (held fixed by its selection rule) . The gain is specific to the true Schwartz ordering: it does not appear for a random permutation or an empirical co-occurrence graph through the identical decoder. A bounded Qwen2 .5-72BInstruct diagnostic shows that supplying the continuum at inference shifts behavior but does not match supervised structured prediction. Theory-aware decoding thus offers a lightweight, controllable way to make value detection faithful to its label space.  \n1 Introduction  \nHuman values underlie moral, social, political, and cultural language, and detecting them in text  \nsupports work across NLP and computational social science. The task is commonly posed as sentence-level multi-label classification: given a sentence, predict which refined human values it expresses (Kiesel et al., 2023 ; Mirzakhmedova et al., 2024) . The dominant modeling approach treats the values as independent labels. This is convenient but theoretically incomplete: the refined Schwartz theory defines the values not as independent categories but as a circular motivational continuum, in which neighboring values are compatible and values on opposite arcs are in tension (Schwartz et al., 2012 ; Schwartz, 2017 ; Schwartz and Cieciuch, 2022) .  \nWe ask whether this theory can be made operational in value detection without overconstraining it. The goal is not a hard rule that opposite values can never co-occur—real texts express conflict, compromise, and trade-offs (Schwartz et al., 2017 ; Skimina et al., 2018)—but a soft inductive bias: can a model preserve predictive performance while producing label sets that are more coherent with the Schwartz continuum? The question is timely. Recent work on the same task shows that strong flat encoders are hard to beat (Ma et al., 2023 ; Molazadeh Oskuee et al., 2023 ; Yeste and Rosso, 2026b), while hard architectural uses of the theory, such as presence gates or higher-order hierarchies, can introduce recall bottlenecks or error propagation (Yeste and Rosso, 2026a) . It is also sharpened by instruction-tuned LLMs, which can be prompted with the theory directly (Sun, 2024 ; Zhu et al., 2025), raising the question of whether supervised structure is still needed.  \nWe encode the 19 refined values as a circular output-space geometry—an angular position per value and a circular distance matrix—and use it in two ways: a training-time penalty and a post-  \nhoc structured decoder over a DeBERTa-v3-base classifier. Across five seeds, training-time geometry yields only limited, non-theory-specific gains, whereas the Schwartz decoder ","cbCaidf34CmvEf4N","https://ap.wps.com/l/cbCaidf34CmvEf4N","pdf",379311,3,1,17,"English","en",105,"# Abstract\n# Introduction\n# Related Work","[{\"question\":\"How does this work move beyond independent-label modeling for Schwartz values?\",\"answer\":\"It represents the 19 refined Schwartz values as a circular output-space geometry, capturing compatibility between neighboring values and tension between opposing arcs.\"},{\"question\":\"What two methods are compared for injecting Schwartz structure into a DeBERTa-v3-base classifier?\",\"answer\":\"The study compares training-time geometry-aware objectives with a post-hoc Schwartz-aware energy decoder that scores whole label sets jointly.\"},{\"question\":\"What is the main result regarding performance and coherence?\",\"answer\":\"Across multiple seeds, training-time geometry yields limited gains, while the Schwartz-aware decoder produces label sets more coherent with the continuum without cost to Macro-F1 or Micro-F1 (as held fixed by its selection rule).\"}]",1784191247,43,{"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},"beyond-independent-labels-schwartz-geometry-decoding-for-human-value-detection","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"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":20},"https://docshare.wps.com/document/research-report/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/beyond-independent-labels-schwartz-geometry-decoding-for-human-value-detection/83889/",4,{"url":51,"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","2026-07-27","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},"How does this work move beyond independent-label modeling for Schwartz values?","Question",{"text":75,"@type":76},"It represents the 19 refined Schwartz values as a circular output-space geometry, capturing compatibility between neighboring values and tension between opposing arcs.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What two methods are compared for injecting Schwartz structure into a DeBERTa-v3-base classifier?",{"text":80,"@type":76},"The study compares training-time geometry-aware objectives with a post-hoc Schwartz-aware energy decoder that scores whole label sets jointly.",{"name":82,"@type":73,"acceptedAnswer":83},"What is the main result regarding performance and coherence?",{"text":84,"@type":76},"Across multiple seeds, training-time geometry yields limited gains, while the Schwartz-aware decoder produces label sets more coherent with the continuum without cost to Macro-F1 or Micro-F1 (as held fixed by its selection rule).","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":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":52,"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"]