[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-514204-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-514204-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","assessing-rule-adherence-of-llm-adjudicators-in-call-of-cthulhu-trpg","Assessing Rule Adherence of LLM Adjudicators in Call of Cthulhu TRPG","","As large language models increasingly act as autonomous adjudicators in tabletop games like Call of Cthulhu, robust rule adherence becomes essential when player intent clashes with system rules. The work identifies Rhetorical Injection, where adversarial users bypass adjudication logic using pseudo-logical reasoning and authoritative coercion. It introduces CoC-Seduce, a multi-agent benchmark with 5,376 samples, and evaluates 22 target adjudicators to measure vulnerability under natural-language interactions.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/assessing-rule-adherence-of-llm-adjudicators-in-call-of-cthulhu-trpg/514204/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/assessing-rule-adherence-of-llm-adjudicators-in-call-of-cthulhu-trpg/514204.png","ImageObject",300,407,{"name":42,"@type":43},"Himbo","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-07","2026-10-06",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",8,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What problem does the paper address for LLM game adjudicators?","Question",{"text":63,"@type":64},"It addresses how reliably LLMs follow explicit game rules when user intent conflicts with the system, especially under naturally phrased, ambiguous player actions.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What is Rhetorical Injection in this context?",{"text":68,"@type":64},"Rhetorical Injection is a manipulation strategy where adversarial users exploit narrative framing, including pseudo-logical reasoning and authoritative coercion, to bypass adjudication logic.",{"name":70,"@type":61,"acceptedAnswer":71},"What is the CoC-Seduce benchmark and how is it used?",{"text":72,"@type":64},"CoC-Seduce is a multi-agent adversarial benchmark built from Call of Cthulhu. It generates 5,376 natural-language samples and benchmarks 22 LLM adjudicators to quantify rule-adherence vulnerability.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},514204,1791413022,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,114,119,122,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":112,"slug":113},30,"research-report",{"id":115,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},9,"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":97,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":55,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":139,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":144,"read_time":145,"preview_page_count":33,"preview_urls":146},687197100911,"https://ap-avatar.wpscdn.com/avatar/a000239b6f1da00475?x-image-process=image/resize,m_fixed,w_180,h_180&k=1785132997149421697","Assessing Rule Adherence of LLM Adjudicators in Call of Cthulhu TRPG  \nWeiying Chen1,3 , Junlong Shen1,3 , Zhanyuan Guo2,3 , Xiaoou Zhou3,*  \n1University of Alberta, Edmonton, AB, Canada  \n2 Qilu University of Technology, Shandong, China  \n3N-Dice Association  \n*Correspondence: [noname512@n-dice.com](noname512@n-dice.com)  \narXiv : 2607 .02802v2 [cs .CL] 2 Oct 2026  \nAbstract  \nAs LLMs are increasingly deployed as autonomous adjudicators in games such as Call of Cthulhu (CoC), robust rule adherence becomes critical when user intent conflicts with system rules. However, as these models are trained to be helpful and compliant, they may be vulnerable to a class of manipulations we term Rhetorical Injection, where adversarial users exploit narrative framing techniques such as pseudo-logical reasoning and authoritative coercion to bypass adjudication logic. We present CoC-Seduce, a multi-agent adversarial benchmark built on CoC, a Tabletop Role-Playing Game (TRPG) in which rules are explicit about which risky actions require adjudication, yet interaction remains entirely in natural language.  \nThree LLMs, i.e., GPT-5.4, Claude Sonnet  \n4.6, Gemini 3.5 Flash, serve as adversarial generators producing 5,376 samples across 4 world settings and 16 skill categories. We then benchmark 22 target adjudicators against this corpus. Evaluation across 22 models reveals that neither newer releases nor explicit reasoning reliably confer adjudication robustness, that PSEUDO-LOGIC framing is the most effective rhetorical style, and that the world setting, including culturally distant ones, has only a modest effect. Project page: [https:](https:)//[github.com/answerrtx/CoC-Seduce](github.com/answerrtx/CoC-Seduce)  \n1 Introduction  \nWith the rapid advancement of Large Language Models (LLMs), there has been a significant surge of interest in deploying AI as autonomous Non-Player Characters (NPCs) or game adjudicators within interactive environments. Among these applications, Tabletop Role-Playing Games (TRPGs), a prominent form of text-based games, have emerged as a natural testing ground (Góngora et al., 2023) for evaluating the logical reasoning and rule-alignment capabilities of these agents with respect to complex underlying mechanics. Unlike  \nFigure 1: A tense moment in a Call of Cthulhu session when the AI Game Master (here serving as Adjudicator) calls for a dice roll—the critical decision point this work evaluates. The image was generated using GPT Image (OpenAI, 2026a) .  \ntraditional video games constrained by fixed choice menus, TRPG-style settings grant players absolute natural language freedom, yet demand that the AI Adjudicator strictly enforces a rigid, underlying rule engine. In this setup, the AI must adjudicate between expressive player intent and the fixed constraints of the underlying rule system. Existing NLP research in this domain has predominantly explored Dungeons & Dragons (D&D) (Krier, 1979) as the primary sandbox (Louis and Sutton, 2018 ; Zhu et al., 2023a ; Delafuente et al., 2025) . However, the player actions captured in these resources are dominated by explicit, combat-focused directives (e.g.,“I cast Fireball\"), for which the need fora dice roll is rarely in question; the adjudication of ambiguous, narrative-framed actions remains largely unexplored.  \nTo push the boundaries of AI adjudication in these text-based domains, we turn to Call of Cthulhu (CoC) (Petersen et al., 2016), a TRPG paradigm that epitomizes highly ambiguous, narrative-driven text games. In CoC, the gameplay shifts from tactical combat to investigative horror, where player inputs are deeply intertwined with atmospheric flavor text, emotional pleas, and de-  \nceptive framing (e.g.,“I take three steps at once, jump directly over the train, then land safely on the other side.\" as shown in Fig. 1) . We hypothesize that, because LLMs are optimized to maintain textual coherence and to please users (Sharma et al., 2025), they tend to treat player inputs a","cbCaiquYfYMXJEFA","https://ap.wps.com/l/cbCaiquYfYMXJEFA","pdf",645670,14,"English","# Abstract\n# Introduction\n# Related Works\n## LLMs for Tabletop Role-Playing Games","[{\"question\":\"What problem does the paper address for LLM game adjudicators?\",\"answer\":\"It addresses how reliably LLMs follow explicit game rules when user intent conflicts with the system, especially under naturally phrased, ambiguous player actions.\"},{\"question\":\"What is Rhetorical Injection in this context?\",\"answer\":\"Rhetorical Injection is a manipulation strategy where adversarial users exploit narrative framing, including pseudo-logical reasoning and authoritative coercion, to bypass adjudication logic.\"},{\"question\":\"What is the CoC-Seduce benchmark and how is it used?\",\"answer\":\"CoC-Seduce is a multi-agent adversarial benchmark built from Call of Cthulhu. It generates 5,376 natural-language samples and benchmarks 22 LLM adjudicators to quantify rule-adherence vulnerability.\"}]","Assessing Rule Adherence of LLM Adjudicators in Call of Cthulhu TRPG | PDF",1791297555,35,[147,151,155,159],{"page_no":22,"url":148,"width":149,"height":150},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/9a7bdbe25365456af14b47b62def71b0.webp",983,1390,{"page_no":26,"url":152,"width":153,"height":154},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/6e7c5346d1071b6e2ec4667fb1046a6f.webp",982,1388,{"page_no":30,"url":156,"width":157,"height":158},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/78258588ad7688cc624a9649f1bb8af8.webp",981,1387,{"page_no":33,"url":160,"width":161,"height":162},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/6de22f03ed79963b15b19d298e204037.webp",980,1385]