[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-427081-105":59,"doc-detail-427081-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","dungeons-and-dragons-as-a-dialog-challenge-for-artificial-intelligence","Dungeons and Dragons - as a Dialog Challenge for Artificial Intelligence","","AI research treats Dungeons and Dragons (D&D) as a benchmark for language capabilities, but frames it here as a dialogue-system challenge. The work defines two tasks: generating the next conversational turn in the game and predicting the game state from dialogue history. A gameplay dataset of nearly 900 games is built, totaling 7,000 players, 800,000 dialogue turns, 500,000 dice rolls, and 58 million words, with automatic partial state annotations. A large language model generates character- and Dungeon Master-style dialogue, evaluated by humans and automatically, including the value of state tracking for producing plausible conversation.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/dungeons-and-dragons-as-a-dialog-challenge-for-artificial-intelligence/427081/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/dungeons-and-dragons-as-a-dialog-challenge-for-artificial-intelligence/427081.png","ImageObject",300,407,{"name":92,"@type":93},"Stanford","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-23","2026-09-29",true,{"@type":102,"interactionType":103,"userInteractionCount":29},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What are the two main tasks in this paper’s Dungeons and Dragons challenge?","Question",{"text":112,"@type":113},"The paper defines Next Utterance Prediction and Game State Tracking. Next Utterance Prediction generates the next game dialogue turn, while Game State Tracking predicts the game state using dialogue history.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How large is the gameplay dataset constructed for this study?",{"text":117,"@type":113},"The dataset covers nearly 900 games with about 7,000 players. It contains roughly 800,000 dialogue turns, 500,000 dice rolls, and 58 million words.",{"name":119,"@type":110,"acceptedAnswer":120},"What kinds of dialogue does the trained language model produce?",{"text":121,"@type":113},"The model can respond either as a specific in-game character or as the Dungeon Master. It is trained to generate in-character roleplaying dialogue and out-of-character discussion of rules or strategy.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},427081,1790811636,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":29,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":144,"read_time":145},2336477552062,"https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d","Dungeons and Dragons as a Dialog Challenge for Artificial Intelligence  \nChris Callison-Burch􀀃  \nUniversity of Pennsylvania [ccb@upenn.edu](ccb@upenn.edu)  \nDaphne Ippolito  \nUniversity of Pennsylvania Google Research  \nGaurav Singh Tomar􀀃  \nGoogle Research [gtomar@google.com](gtomar@google.com)  \nLara J. Martin  \nUniversity of Pennsylvania  \nSuma Bailis  \nGoogle Research  \nDavid Reitter  \nGoogle Research  \nAbstract  \nAI researchers have posited Dungeons and Dragons (D&D) as a challenge problem to test systems on various language-related capabilities. In this paper, we frame D&D specifically as a dialogue system challenge, where the tasks are to both generate the next conversational turn in the game and predict the state of the game given the dialogue history. We create a gameplay dataset consisting of nearly 900 games, with a total of 7,000 players, 800,000 dialogue turns, 500,000 dice rolls, and 58 million words. We automatically annotate the data with partial state information about the game play. We train a large language model (LM) to generate the next game turn, conditioning it on different information. The LM can respond as a particular character or as the player who runs the game—i.e., the Dungeon Master (DM) . It is trained to produce dialogue that is either in-character (roleplaying in the fictional world) or out-of-character (discussing rules or strategy) . We perform a human evaluation to determine what factors make the generated output plausible and interesting. We further perform an automatic evaluation to determine how well the model can predict the game state given the history and examine how well tracking the game state improves its ability to produce plausible conversational output.  \n1 Introduction  \nArtificial Intelligence has a long and rich history of using games as challenge problems that lead to advances in the field. In many cases, AI game-playing systems have gone on to rival human champions of the game. Dungeons and Dragons has been identified as an appropriate challenge for the next stage of artificial intelligence (Ellis and Hendler, 2017 ; Louis and Sutton, 2018 ; Martin et al., 2018b) . Ellis and Hendler (2017) proposed open-ended creative games like D&D as the next challenge for AI after the human-level successes of AI at Chess and  \nDenotes equal contribution  \nGo, which are zero-sum, deterministic, sequential two-player games with perfect information. Louis and Sutton (2018) understood the importance of narrative in natural language processing (NLP) and generation (NLG) . In particular, they saw how cooperative story generation between humans already exists in these games and can be used for automated generation. Martin et al. (2018b) outlined some of the specific challenges D&D presents to the NLP community; such as a state of the game world distributed across the Dungeon Master (DM) and other players or dealing with the intrinsic rewards players get from taking certain actions that would not necessarily provide them with points in the game.  \nD&D involves multiple players who roleplay characters in a fantasy setting, guided by a Dungeon Master who sets obstacles and adventures and plays as monsters. In roleplaying games like Dungeons and Dragons, the gameplay happens through language rather than moves on a game board, making it an interesting domain for NLP research. To have an AI successfully play D&D , it would require abilities like  \n• Language generation (multi-party dialog, generating descriptions of the world/actions, storytelling)  \n• Language understanding (knowledge acquisition and representation, state tracking, automated reasoning)  \n• Planning / strategic play during battles (similar to chess or go)  \nAppendix A gives an example of D&D gameplay and the AI challenges presented by it.  \nIs it possible to design an AI system that is capable of playing a game of D&D either as a character in the game or as the Dungeon Master using current AI technology? We argue that now is the perfect time","cbCaiaEWg3cO1wiJ","https://ap.wps.com/l/cbCaiaEWg3cO1wiJ","pdf",757133,15,"English","# Abstract\n# 1 Introduction\n# 2 Tasks\n## Next Utterance Prediction\n## Game State Tracking","[{\"question\":\"What are the two main tasks in this paper’s Dungeons and Dragons challenge?\",\"answer\":\"The paper defines Next Utterance Prediction and Game State Tracking. Next Utterance Prediction generates the next game dialogue turn, while Game State Tracking predicts the game state using dialogue history.\"},{\"question\":\"How large is the gameplay dataset constructed for this study?\",\"answer\":\"The dataset covers nearly 900 games with about 7,000 players. It contains roughly 800,000 dialogue turns, 500,000 dice rolls, and 58 million words.\"},{\"question\":\"What kinds of dialogue does the trained language model produce?\",\"answer\":\"The model can respond either as a specific in-game character or as the Dungeon Master. It is trained to generate in-character roleplaying dialogue and out-of-character discussion of rules or strategy.\"}]","Dungeons and Dragons - as a Dialog Challenge for Artificial Intelligence | PDF",1790642741,38]