[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83842-en":3,"doc-seo-83842-105":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":21,"is_downloadable":21,"audit_status":21,"page_count":20,"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},83842,2336464648322,"Aria","https://ap-avatar.wpscdn.com/avatar/2200025388227c56fec?_k=1778556882303663488",8,"Research & Report","PAST-TIDE Prototype Anchored Statement Tuning with Topic Invariant Normalization for Stance Detection","PAST-TIDE presents a stance detection system tackling both subtasks of the StanceNakba Shared Task at NakbaNLP@LREC-COLING 2026 using statement tuning. Stance is redefined as cloze-style masked language modeling, where a multi-token verbalizer maps label words via the pre-trained MLM head rather than adding a new classifier head. Prototypical contrastive learning with persistent class prototypes supports small batches, and topic-conditional layer normalization enables cross-topic Arabic transfer, achieving macro-F1 scores of 0.75 and 0.74 on the official leaderboard.","arXiv :2607 .04690v 1 [ cs .CL] 6 Jul 2026  \nPAST-TIDE: Prototype-Anchored Statement Tuning with Topic-Invariant Normalization for Stance Detection  \nMd. Shakhoyat Rahman Shujon 1 , MD Jahid Hasan Jim 1 , Md. Milon Islam 1 ,  \nMd. Rezwanul Haque2 , Fakhri Karray2 ,3  \n1 Department of Computer Science and Engineering, Khulna University of Engineering & Technology [skt104.shujon@gmail.com](skt104.shujon@gmail.com) , [mdjahidhasanjim277@gmail.com](mdjahidhasanjim277@gmail.com) ,  \n[milonislam@cse.kuet.ac.bd](milonislam@cse.kuet.ac.bd)  \n2 Department of Electrical and Computer Engineering, University of Waterloo  \n3 Department of Machine Learning, Mohamed bin Zayed University of Artificial Intelligence  \n{rezwan, [karray}@uwaterloo.ca](karray}@uwaterloo.ca)  \nAbstract  \nWe introduce PAST-TIDE, our stance detection system addressing both subtasks of the StanceNakba Shared Task at NakbaNLP@LREC-COLING 2026 . The main idea is statement tuning. We redefine stance as cloze-style masked language modeling (MLM), letting a verbalizer map label words to stance categories through the pre-trained MLM head rather than appending a randomly initialized classification head. We complement this with prototypical contrastive learning, which uses learnable class prototypes for batch-size independent contrastive training, and topic-conditional layer normalization for cross-topic Arabic stance detection. PAST-TIDE achieves macro-F1 scores of 0.75 for Subtask A and 0.74 for Subtask B on the official leaderboard, indicating that minimal architectural additions to a pre-trained model can remain competitive in low-resource settings.  \nKeywords: stance detection, prompt-based classification, contrastive learning, cross-topic generalization, low-resource multilingual NLP.  \n1. Introduction  \nUnderstanding a person’s stance on a given topic remains a challenging problem in computational argumentation (Mohammad et al. , 2016 ; Küçük and Can , 2020) . The field has moved from handcrafted features to pre-trained language model finetuning (Ghosh et al. , 2019 ; Li and Caragea, 2021) to prompt-based reformulations (Schick and Schütze, 2021 ; Gao et al. , 2021), yet multilingual stance detection remains underexplored. Most of the existing work focuses on benchmarks in English alone with thousands of labeled samples, leaving open the question of whether prompt-based methods remain effective when labels are limited and the target language is not English. The eNakba Shared Task (Aldous et al. , 2026) addresses exactly the above gap. It defines two subtasks: Subtask A, stance at the actor-level in English (Pro-Palestine/ProIsrael/Neutral), and Subtask B, stance cross-topic in Arabic (Favor/Against/ None) . Both tasks use relatively small datasets, with 1,401 and 1,205 labeled samples, split into 70/15/15 for train, development, and test. A standard [CLS] classifier head lacks any pre-trained initialization of its own, and that  \nProceedings of the 2nd International Workshop on Nakba Narratives as Language Resources (Nakba-NLP 2026) @ LREC 2026, pages 252–257  \n11 May 2026. ©ELRA Language Resources Association (ELRA), 2026  \npre-training/fine-tuning gap degrades performance at this scale.  \nOur architecture, PAST-TIDE, is based on one idea: the classistance classification capacity already exists withinined MLM head, and the task is to access it rather than replace it. The major contributions of this paper are as follows. First, statement tuning redefines stance as cloze-style MLM, bypassing the randomly initialized [CLS] head with a multi-token verbalizer that averages predictions across register-diverse label words. Second, prototypical contrastive learning (PCL) replaces inbatch negatives with K =3 class prototypes, where these prototypes are updated by gradients and remain persistent, giving stable anchors even at small batch sizes. Third, topic-conditional layer normalization (T-CLN), inspired by conditional normalization in vision, generates per-topic normalizatio","cbCaitUw083wzNTb","https://ap.wps.com/l/cbCaitUw083wzNTb","pdf",328877,6,1,"English","en",105,"# Introduction\n# Related Work","[{\"question\":\"What is PAST-TIDE’s core idea for stance detection?\",\"answer\":\"PAST-TIDE relies on statement tuning by redefining stance as cloze-style masked language modeling, using the pre-trained MLM head with a verbalizer instead of an added classifier head.\"},{\"question\":\"How does PAST-TIDE handle small-batch contrastive training?\",\"answer\":\"It uses prototypical contrastive learning with learnable class prototypes (K=3), replacing in-batch negatives and maintaining stable anchors even with small batch sizes.\"},{\"question\":\"How does the system support cross-topic stance detection in Arabic?\",\"answer\":\"It introduces topic-conditional layer normalization (T-CLN), generating per-topic normalization parameters to improve transfer across topics.\"}]",1784190919,15,{"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},"past-tide-prototype-anchored-statement-tuning-with-topic-invariant-normalization-for-stance-detection","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":21},"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/past-tide-prototype-anchored-statement-tuning-with-topic-invariant-normalization-for-stance-detection/83842/",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","2026-07-26","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 is PAST-TIDE’s core idea for stance detection?","Question",{"text":75,"@type":76},"PAST-TIDE relies on statement tuning by redefining stance as cloze-style masked language modeling, using the pre-trained MLM head with a verbalizer instead of an added classifier head.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does PAST-TIDE handle small-batch contrastive training?",{"text":80,"@type":76},"It uses prototypical contrastive learning with learnable class prototypes (K=3), replacing in-batch negatives and maintaining stable anchors even with small batch sizes.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the system support cross-topic stance detection in Arabic?",{"text":84,"@type":76},"It introduces topic-conditional layer normalization (T-CLN), generating per-topic normalization parameters to improve transfer across topics.","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},[93,97,101,105,110,114,119,122,127,130,134],{"id":21,"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":20,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]