[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85513-en":3,"doc-seo-85513-105":28,"detail-sidebar-cat-0-en-105":90},{"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":20,"is_downloadable":20,"audit_status":20,"page_count":11,"language":21,"language_code":22,"site_id":23,"html_lang":22,"table_of_contents":24,"faqs":25,"seo_title":13,"seo_description":14,"update_tm":26,"read_time":27},85513,34359740700684,"Finn","https://ap-avatar.wpscdn.com/avatar/1f400023980c374ae676?_k=1777273430885731487",8,"Research & Report","The Case of the Mysterious Citations","Mysterious citations—references whose cited titles cannot be verified at the claimed location—are increasingly appearing in peer-reviewed scientific venues. The paper presents a semi-automated pipeline to analyze papers from four major high-performance computing conferences in 2021 and 2025, before and after widespread generative-AI adoption. It finds zero mysterious citations and incorrect author attributions in 2021, but multiple in every 2025 proceeding, totaling 38 mysterious citations and 46 incorrect-author cases. Conference policies requiring AI disclosure were not acknowledged by any paper.","The Case of the Mysterious Citations  \nAmanda Bienz, University of New Mexico, Albuquerque, NM, USA Carl Pearson, Sandia National Laboratories, Albuquerque, NM, USA  \nSimon Garcia de Gonzalo, Sandia National Laboratories, Albuquerque, NM, USA  \narXiv :2602 .05867v2 [ cs .DL] 11 Jul 2026  \nAbstract—Mysterious citations are routinely appearing in peer-reviewed publications throughout the scientific community. In this paper, we developed a semi-automated pipeline and examine the proceedings of four major high-performance computing conferences for both 2021 and 2025, pre- and post-generative AI respectively. We analyzed each paper for mysterious citations, or those in which the cited title was verified to not exist at the cited location. Further, we analyzed author correctness throughout citations, flagging citations for which more than half of the cited authors did not match those listed on the cited publication. While there were zero cases of mysterious citations or incorrect author attributions in any 2021 publication, there were multiple of each in every 2025 conference. In total, there were 38 mysterious citations and 46 citations for which the majority of listed authors incorrect, spanning all four 2025 proceedings. No paper within our dataset acknowledged using AI to generate citations even though all four conference policies  \nrequired acknowledging AI usage, indicating current policies are insufficient.  \nL  \narge language models (LLMs) have transitioned in just a few years from research prototypes to widely available tools. Systems such as Chat-  \nGPT [1], Claude [2], and Gemini [3] are now embedded in everyday workflows, offering fluent text generation, summarization, and synthesis at low cost. Accessibility, ease of use, and apparent competence have made them attractive assistants for writing tasks in various domains, including academic research [4] .  \nThis shift has begun to surface explicitly in the academic publication ecosystem. Major venues and professional societies, including ACM and IEEE, now publish guidelines on the use of generative AI in paper preparation[5], [6], ranging from disclosure to authorship attribution. These policies acknowledge a reality: LLMs are already being used in conference and journal submissions, particularly for drafting, editing, and literature review assistance, and their presence is no longer exceptional.  \nAt the same time, the use of LLMs in publications is often not transparent. Many venues do not require explicit disclosure, and even when policies exist, compliance is uneven. Anecdotal reports from program committees and reviewers increasingly describe a specific failure mode: “hallucinated” or outright fabricated citations inserted by LLMs for related work sections or to substantiate claims made in the publication.  \nThese references often appear superficially plausible, complete with author lists (featuring names of actual researchers), real venues, and publication dates, but correspond to no real publication.  \nThis phenomenon poses a burden on the peerreview process. Reviewers cannot reliably detect such errors by inspection alone. Verifying a suspicious reference may require checking reference data (e.g. titles, author names, venues, DOIs) against external publication records (libraries, metadata aggregators), an effort that scales poorly when combined with an already time-consuming review workflow. As a result, fabricated or hallucinated citations can pass unnoticed, undermining the integrity of the evaluation process.  \nThis paper outlines an effort to make this emerging issue visible. By analyzing papers published in computer science conferences, we quantify the prevalence of \"mysterious\" citations, or those in which the cited title is not found at the cited location and also cannot be located through Google searches. Rather than treating these incidents as isolated anecdotes, we provide empirical evidence of their scope and correlation to the advent of large language m","cbCaimos9NAzDl7b","https://ap.wps.com/l/cbCaimos9NAzDl7b","pdf",251942,1,"English","en",105,"# Abstract\n# Use and Misuse of LLMs in Academic Writing\n## Hallucinated or Fabricated References\n## Impact on Peer Review Verification\n## Quantifying Prevalence Across Conference Proceedings\n## Policy and AI Disclosure Gaps","[{\"question\":\"What are “mysterious citations” in the paper’s analysis?\",\"answer\":\"They are citations where the cited title is verified to not exist at the stated location and cannot be found through the described searches.\"},{\"question\":\"How did the authors measure the problem across conferences?\",\"answer\":\"They built a semi-automated pipeline and analyzed papers from four major high-performance computing conferences in 2021 and 2025, then checked cited titles and the correctness of cited author lists.\"},{\"question\":\"What did the study find when comparing 2021 to 2025?\",\"answer\":\"2021 showed zero cases of mysterious citations or incorrect author attributions, while 2025 showed multiple such cases in every proceeding, totaling 38 mysterious citations and 46 citations with mostly incorrect authors.\"}]",1784204105,20,{"code":4,"msg":29,"data":30},"ok",{"site_id":23,"language":22,"slug":31,"title":13,"keywords":32,"description":14,"schema_data":33,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":26},"the-case-of-the-mysterious-citations","",{"@graph":34,"@context":84},[35,52,67],{"@type":36,"itemListElement":37},"BreadcrumbList",[38,42,46,49],{"item":39,"name":40,"@type":41,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":43,"name":44,"@type":41,"position":45},"https://docshare.wps.com/document/","Document",2,{"item":47,"name":12,"@type":41,"position":48},"https://docshare.wps.com/document/research-report/",3,{"item":50,"name":13,"@type":41,"position":51},"https://docshare.wps.com/document/the-case-of-the-mysterious-citations/85513/",4,{"url":50,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":22,"description":14,"dateModified":60,"datePublished":61,"encodingFormat":59,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":39,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":20},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What are “mysterious citations” in the paper’s analysis?","Question",{"text":74,"@type":75},"They are citations where the cited title is verified to not exist at the stated location and cannot be found through the described searches.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How did the authors measure the problem across conferences?",{"text":79,"@type":75},"They built a semi-automated pipeline and analyzed papers from four major high-performance computing conferences in 2021 and 2025, then checked cited titles and the correctness of cited author lists.",{"name":81,"@type":72,"acceptedAnswer":82},"What did the study find when comparing 2021 to 2025?",{"text":83,"@type":75},"2021 showed zero cases of mysterious citations or incorrect author attributions, while 2025 showed multiple such cases in every proceeding, totaling 38 mysterious citations and 46 citations with mostly incorrect authors.","https://schema.org",{"og:url":50,"og:type":86,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":88,"canonical":50},"index,follow",{"doc_id":7,"site_id":23},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,126,129,133],{"id":20,"doc_module":4,"doc_module_name":44,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":45,"doc_module":4,"doc_module_name":44,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":51,"doc_module":4,"doc_module_name":44,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":44,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":44,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":44,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":44,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":44,"category_name":124,"show_sort_weight":27,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":27,"doc_module":4,"doc_module_name":44,"category_name":127,"show_sort_weight":27,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":44,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":44,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]