[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-503828-105":59,"doc-detail-503828-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","rank-list-sensitivity-of-recommender-systems-to-interaction-perturbations","Rank List Sensitivity of Recommender Systems to Interaction Perturbations","","Prediction models may become unstable when training data changes, leading to conflicting test-time outputs for individual points. This study examines such sensitivity in recommender systems, where minor perturbations in unrelated users’ interactions can drastically alter everyone’s recommendation lists. The paper defines Rank List Sensitivity (RLS) to quantify how test-time rank lists change under training-data perturbations and proposes CASPER to induce minimal, systematic perturbations. Experiments on four datasets show extreme sensitivity to random and CASPER-induced changes, with greater instability for low-accuracy users than for high-accuracy users.",{"@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/rank-list-sensitivity-of-recommender-systems-to-interaction-perturbations/503828/",{"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/rank-list-sensitivity-of-recommender-systems-to-interaction-perturbations/503828.png","ImageObject",300,407,{"name":92,"@type":93},"hdkawk","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-17","2026-10-03",true,{"@type":102,"interactionType":103,"userInteractionCount":39},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does the paper study in recommender systems?","Question",{"text":112,"@type":113},"It studies how recommender models can be sensitive to small changes in training data, where perturbations in other users’ interactions can drastically change recommendation outputs for test-time users.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How is stability measured in this work?",{"text":117,"@type":113},"The paper introduces Rank List Sensitivity (RLS), which measures how much rank lists generated at test time change in response to perturbations in the training data.",{"name":119,"@type":110,"acceptedAnswer":120},"What is CASPER and what does it do?",{"text":121,"@type":113},"CASPER is a method that uses cascading effects to identify minimal and systematic training-data perturbations that induce higher instability in recommender systems.","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},503828,1791418622,{"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":39,"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},3985747859855,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","Rank List Sensitivity of Recommender Systems to Interaction Perturbations  \nSejoon Oh  \n[soh337@gatech.edu](soh337@gatech.edu)[ ](soh337@gatech.edu)Georgia Institute of Technology United States  \nBerk Ustun  \n[berk@ucsd.edu](berk@ucsd.edu)[ ](berk@ucsd.edu)University of California San Diego  \nUnited States  \nJulian McAuley  \n[jmcauley@eng.ucsd.edu](jmcauley@eng.ucsd.edu)[ ](jmcauley@eng.ucsd.edu)University of California San Diego United States  \nSrijan Kumar  \n[srijan@gatech.edu](srijan@gatech.edu)[ ](srijan@gatech.edu)Georgia Institute of Technology United States  \nABSTRACT  \nPrediction models can exhibit sensitivity with respect to training data: small changes in the training data can produce models that assign conflicting predictions to individual data points during test time. In this work, we study this sensitivity in recommender systems, where users’ recommendations are drastically altered by minor perturbations in other unrelated users’ interactions. We introduce a measure of stability for recommender systems, called Rank List Sensitivity (RLS), which measures how rank lists generated by a given recommender system at test time change as a result of a perturbation in the training data. We develop a method, CASPER, which uses cascading effect to identify the minimal and systematical perturbation to induce higher instability in a recommender system. Experiments on four datasets show that recommender models are overly sensitive to minor perturbations introduced randomly or via CASPER — even perturbing one random interaction of one user drastically changes the recommendation lists of all users. Importantly, with CASPER perturbation, the models generate more unstable recommendations for low-accuracy users (i.e., those who receive low-quality recommendations) than high-accuracy ones.  \nCCS CONCEPTS  \n• Information systems → Recommender systems.  \nKEYWORDS  \nRecommender Systems, Model Stability, Input Data Perturbation  \nACM Reference Format:  \nSejoon Oh, Berk Ustun, Julian McAuley, and Srijan Kumar. 2022. Rank List Sensitivity of Recommender Systems to Interaction Perturbations. In Proceedings of the 31st ACM International Conference on Information and Knowledge Management (CIKM’22), October 17–21, 2022, Atlanta, GA, USA. ACM, New York, NY, USA, 11 pages. [https://doi.org/10.1145/3511808.3557425](https://doi.org/10.1145/3511808.3557425)  \nPermission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions [from permissions@acm.org](from permissions@acm.org).  \nCIKM’22, October 17–21, 2022, Atlanta, GA, USA © 2022 Association for Computing Machinery.  \nACM ISBN 978-1-4503-9236-5/22/10. . . $15.00 [https://doi.org/10.1145/3511808.3557425](https://doi.org/10.1145/3511808.3557425)  \nFigure 1: Small changes (e.g., leave-one-out perturbation) in the training data can produce recommender systems that output drastically different recommendations for all individual users.  \n1 INTRODUCTION  \nSmall changes in training data can produce large changes in outputs of machine learning models [7, 40, 47, 57, 59, 75] . Marx et al. [45] showed how classification tasks can often admit competing models that perform almost equally well in terms of an aggregate performance metric (e.g., error rate, AUC) but that assign conflicting predictions to individual data points. Likewise, Black and Fredrikson [6] showed how removing a point from a training dataset can produce models that assign drastically different predictions, and highlighted how this lack of stability disproportionatel","cbCaiaMDUv1Fholk","https://ap.wps.com/l/cbCaiaMDUv1Fholk","pdf",2202381,11,"English","# Abstract\n# Introduction\n## Sensitivity of machine learning outputs to training changes\n## Implications for recommender systems and real-world domains\n# Method Overview\n## Rank List Sensitivity (RLS)\n## CASPER perturbation approach\n# Experiments\n## Dataset results and stability comparisons\n# Conclusion","[{\"question\":\"What problem does the paper study in recommender systems?\",\"answer\":\"It studies how recommender models can be sensitive to small changes in training data, where perturbations in other users’ interactions can drastically change recommendation outputs for test-time users.\"},{\"question\":\"How is stability measured in this work?\",\"answer\":\"The paper introduces Rank List Sensitivity (RLS), which measures how much rank lists generated at test time change in response to perturbations in the training data.\"},{\"question\":\"What is CASPER and what does it do?\",\"answer\":\"CASPER is a method that uses cascading effects to identify minimal and systematic training-data perturbations that induce higher instability in recommender systems.\"}]","Rank List Sensitivity of Recommender Systems to Interaction Perturbations | PDF",1791060841,28]