[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83870-en":3,"doc-seo-83870-105":30,"detail-sidebar-cat-0-en-105":92},{"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},83870,8796095461564,"Liam","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","You Frame It How Conceptual Representations Shape LLM Detection and Reasoning about Antisemitism","Large language models can incorporate external conceptual resources at inference time, enabling new approaches to detecting and explaining ideologically and historically complex phenomena such as antisemitism. The study evaluates how definitional, fine-grained taxonomic, example-augmented, and large-context representations influence antisemitism detection across four leading LLMs using two expert-annotated datasets. Fine-grained taxonomies improve recall but reduce precision, while larger resources provide no extra quantitative gains; post-Holocaust antisemitism remains hardest. Explanation analysis identifies systematic reference overproduction, lexical cue reliance, overconfidence, and difficulty with subtle justificatory forms.","You Frame It: How Conceptual Representations Shape LLM Detection and  \nReasoning about Antisemitism  \nKatharina Soemer  \nGoethe University, Frankfurt, Germany [soemer@soz.uni-frankfurt.de](soemer@soz.uni-frankfurt.de)  \nHelena Mihaljevi  \nHTW Berlin, Germany [mihalje@htw-berlin.de](mihalje@htw-berlin.de)  \narXiv :2607 .04945v 1 [ cs .CL] 6 Jul 2026  \nAbstract  \nLLMs enable the integration of external conceptual resources at inference time, creating new opportunities for detecting ideologically and historically complex phenomena such as antisemitism. We investigate how different forms of conceptual grounding affect antisemitism detection and explanation behavior across four state-of-the-art LLMs. Using two expert-annotated datasets, we compare definitional, fine-grained taxonomic, exampleaugmented, and large-context representations of antisemitism. We find that fine-grained taxonomic representations substantially improve recall, while simultaneously reducing precision. Surprisingly, supplying substantially larger conceptual resources yields no additional quantitative benefit. Post-Holocaust antisemitism poses the most persistent challenge across models and configurations. Analysis of explanations further reveals systematic limitations including overproduction of conceptual references, reliance on lexical cues, overconfidence, and difficulties with subtle or justificatory forms of antisemitism. Our findings highlight both the potential and the remaining limitations of conceptually grounded LLMs for antisemitism detection and reasoning.  \n1 Introduction  \nDespite advances in hate speech detection using Large Language Models (LLMs) (Gilardi et al., 2023 ; Li et al., 2024), recent studies indicate that antisemitic content detection remains particularly challenging (Steffen et al., 2024 ; Patel et al., 2025) . Models struggle particularly with implicit, coded, or trope-based rhetoric (Mihaljevi and Steffen, 2022 ; Mendelsohn et al., 2023), and may reproduce or amplify anti-Jewish biases (Anti-Defamation League, 2025) .  \nAntisemitism poses a particular challenge for automated detection due to its conceptual and rhetor-  \nical complexity.1 It operates through a heterogeneous, historically evolving repertoire of stereotypes, conspiracy theories, and narratives which are not always explicitly hateful, e.g., claims of Jewish financial power and control (Becker et al., 2024b) . Even expert annotators show substantial disagreement (Steffen et al., 2023), indicating that LLMs require specific contextual knowledge to capture these ideological patterns.  \nThe knowledge encoded in LLMs is opaque, fixed at training time, and potentially shaped by the same biases the models are expected to detect. Prompting offers the possibility to inject external conceptual representations at inference time, for example through definitions, taxonomies, or examples. While external knowledge infusion has been explored in adjacent areas (Yang et al., 2025 ; Melis et al., 2025), little is known about how different forms of conceptual grounding affect detection and classification of different forms of antisemitism.  \nIn this work, we systematically evaluate the effect of external conceptual representations on antisemitism detection and classification across four state-of-the-art LLMs: Gemini-2.5-Flash, Claude Sonnet 4 .6, GPT-5 .4, and LLaMA-3 .3-70b-instruct. We compare four prompting configurations: a baseline without external knowledge; the IHRA Working Definition of Antisemitism (hereafter: IHRA)(International Holocaust Remembrance Alliance, 2016); and a fine-grained taxonomy derived from the Decoding Antisemitism Lexicon (hereafter: Lexicon) (Becker et al., 2024b) with and without examples. In addition, we evaluate Gemini using the complete 550-page Lexicon.  \nOur study makes four main contributions: (1) We provide the first systematic multi-model evaluation of external conceptual grounding for antisemitism detection and classification. (2) We show that com- ","cbCaiuc3T7IS1mTR","https://ap.wps.com/l/cbCaiuc3T7IS1mTR","pdf",389514,6,1,17,"English","en",105,"# Introduction\n# Related Work","[{\"question\":\"How does external conceptual grounding affect antisemitism detection in the study?\",\"answer\":\"Different conceptual grounding methods are compared, including the IHRA definition, fine-grained taxonomies, and example-augmented prompts. Fine-grained taxonomic representations significantly improve recall, though they reduce precision.\"},{\"question\":\"Which antisemitism type is most challenging for the models?\",\"answer\":\"Post-Holocaust antisemitism poses the most persistent challenge across models and prompting configurations.\"},{\"question\":\"What limitations emerge from analyzing the models’ explanations?\",\"answer\":\"The analysis shows systematic overproduction of conceptual references, reliance on lexical cues, overconfidence, and difficulties handling subtle or justificatory forms of antisemitism.\"}]",1784191104,43,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"you-frame-it-how-conceptual-representations-shape-llm-detection-and-reasoning-about-antisemitism","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"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":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/you-frame-it-how-conceptual-representations-shape-llm-detection-and-reasoning-about-antisemitism/83870/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-07-25","2026-07-16",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"How does external conceptual grounding affect antisemitism detection in the study?","Question",{"text":76,"@type":77},"Different conceptual grounding methods are compared, including the IHRA definition, fine-grained taxonomies, and example-augmented prompts. Fine-grained taxonomic representations significantly improve recall, though they reduce precision.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"Which antisemitism type is most challenging for the models?",{"text":81,"@type":77},"Post-Holocaust antisemitism poses the most persistent challenge across models and prompting configurations.",{"name":83,"@type":74,"acceptedAnswer":84},"What limitations emerge from analyzing the models’ explanations?",{"text":85,"@type":77},"The analysis shows systematic overproduction of conceptual references, reliance on lexical cues, overconfidence, and difficulties handling subtle or justificatory forms of antisemitism.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},"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":107,"slug":138},19,"General","general"]