[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86314-en":3,"doc-seo-86314-105":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},86314,13056703020460,"Valentina","https://ap-avatar.wpscdn.com/avatar/be000253dac470eee5d?_k=1778207105932848923",8,"Research & Report","MET: Theory-Grounded and Culture-Aware Multilingual Moral Reasoning","Language models increasingly support moral decision-making across linguistic and cultural contexts, but prior research misses multilinguality in evaluation design, inference prompting, and training supervision. This work introduces MCLASH, a culturally adapted multilingual moral benchmark capturing situated intuitions and norms. It also proposes MET, a theory-grounded two-step prompting framework that selects situation- and culture-specific grounds before reasoning in the user’s language. MET-D further distills the second step without external supervision, improving macro-F1 and native-language reasoning gains across multiple model families.","arXiv :2607 . 1 1736v 1 [ cs .CL] 13 Jul 2026  \nMET: Theory-Grounded and Culture-Aware Multilingual Moral Reasoning  \nAyoung Lee1, Ryan Kwon1, Yunxiang Zhang1, Yuxuan Liu1, Peter Railton2, Lu Wang1 1Department of Computer Science and Engineering  \n2Department of Philosophy University of Michigan Ann Arbor, MI, USA  \n{leeay, ryankwon, yunxiang, yuxualiu, prailton, [wangluxy](wangluxy}@umich.edu)[}](wangluxy}@umich.edu)[@umich.edu](wangluxy}@umich.edu)[ ](wangluxy}@umich.edu) [huggingface.co/collections/launch/met](huggingface.co/collections/launch/met)  github.com/aylee2008/met  \nAbstract  \nLanguage models are increasingly used for moral decision-making across diverse linguistic and cultural contexts, yet existing work overlooks multilinguality on three aspects: 1) multilingual evaluation benchmarks use direct translation, failing to adapt culture-specific items; 2) inference-time methods for moral reasoning rely on static, English-centric scaffolds and lack grounding in moral theory; 3) training methods for moral decisionmaking typically require expensive supervision from stronger models or human annotators. We address these gaps with three contributions. First, we introduce MCLASH, a multilingual moral decision-making benchmark to capture culturally situated moral intuitions and social norms across languages. Second, we propose MET (Multilingual Ethics with Theorygrounded reasoning), a two-step prompting method built on expert-curated, theory-based grounds drawn from psychology and philosophy: the model first selects situation-and culture-specific grounds, then reasons over them in the native language of the user. Third, we introduce MET-D (METDistillation), which enhances the second step through a self-distillation training stage that requires no external supervision. MET-D improves macro-F1 over the base model on all three models of different sizes and families (Qwen3-4B, Qwen3-8B, Gemma3-4B), by an average of 3.71 points on MCLASH and 4.23 on MMoralExceptQA, with a peak MCLASH gain of 12.94 points for Malay on Qwen3-8B. We further reveal that MET-D increases native-language reasoning by 62.13 points on average, and that beneficial grounds differ systematically across cultures. Together, these contributions open the path for culture-aligned, theory-grounded multilingual moral reasoning.  \n1 Introduction  \nLanguage models are increasingly deployed in moral decision-making across high-stakes domains such as law (Xiao et al., 2021; Nguyen, 2023) and healthcare (Singhal et al., 2023; Hu et al., 2024), where users span diverse cultural backgrounds. Effective deployment in such settings requires that models reason within the linguistic and cultural context of the user, accounting for the local value systems that shape a sound moral judgment. However, culturally aligned moral decision-making remains overlooked in both evaluation and methods.  \nEvaluation: direct translation strips cultural context from moral dilemmas. Existing multilingual moral benchmarks directly translate English scenarios (Kumar & Jurgens, 2025; Farid et al., 2025) without adapting culture-specific items (Newmark, 1988) such as named entities, institutions, currencies, and customs. When directly translated into Korean or Hindi, a U.S. scenario set during “Thanksgiving” or at a “Wendy’s” can import improper  \n: Dimensions : Grounds : Descriptions  \nFigure 1: Example ground selection in different languages:  Grounds are organized into  dimensions and paired with descriptions . For the same situation, the selected grounds vary by language: in English, First-principles reasoning reflects the Western rationalism, whereas in Korean, Contractarianism reflects Confucian values.  \nWestern cultural context into account. We address this with MCLASH, a multilingual moral decision-making benchmark built by culturally adapting CLASH (Lee et al., 2025)—a corpus of long-form high-stakes scenarios paired with character descriptions—into five additional languages (Chinese, Hind","cbCaig3mEwMtryIY","https://ap.wps.com/l/cbCaig3mEwMtryIY","pdf",2655287,2,1,31,"English","en",105,"# Introduction\n## Evaluation\n## Method (inference)\n## Method (training)","[{\"question\":\"What problem does this paper address in multilingual moral decision-making?\",\"answer\":\"It addresses the lack of culturally aligned multilingual evaluation and methods: benchmarks often rely on direct translation, inference prompting lacks theory grounding and culture adaptation, and training methods typically require expensive supervision.\"},{\"question\":\"What is MCLASH and what does it measure?\",\"answer\":\"MCLASH is a multilingual moral decision-making benchmark built by culturally adapting CLASH scenarios into multiple languages. It captures culturally situated moral intuitions and social norms with situation-level descriptions and paired character information.\"},{\"question\":\"How does MET work at inference time?\",\"answer\":\"MET is a two-step prompting method: the model first selects situation- and culture-specific moral grounds, then generates a grounded reasoning chain in the native language of the user using those grounds.\"}]",1784210421,78,{"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},"met-theory-grounded-and-culture-aware-multilingual-moral-reasoning","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":20},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/met-theory-grounded-and-culture-aware-multilingual-moral-reasoning/86314/",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},"What problem does this paper address in multilingual moral decision-making?","Question",{"text":75,"@type":76},"It addresses the lack of culturally aligned multilingual evaluation and methods: benchmarks often rely on direct translation, inference prompting lacks theory grounding and culture adaptation, and training methods typically require expensive supervision.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What is MCLASH and what does it measure?",{"text":80,"@type":76},"MCLASH is a multilingual moral decision-making benchmark built by culturally adapting CLASH scenarios into multiple languages. It captures culturally situated moral intuitions and social norms with situation-level descriptions and paired character information.",{"name":82,"@type":73,"acceptedAnswer":83},"How does MET work at inference time?",{"text":84,"@type":76},"MET is a two-step prompting method: the model first selects situation- and culture-specific moral grounds, then generates a grounded reasoning chain in the native language of the user using those grounds.","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":20,"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"]