[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82298-en":3,"doc-seo-82298-105":29,"detail-sidebar-cat-0-en-105":91},{"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":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},82298,13056703019662,"Evangeline","https://ap-avatar.wpscdn.com/avatar/be000253a8e92610077?_k=1778726343310543188",8,"Research & Report","Creativity, honesty and designed forgetting emerge in small hyperbolic language models","Language models are optimized for scale, yet function rather than companion. When an assistant personalizes and accumulates user-specific memory, it can quietly become someone and acquire traits that harm the user. Trained human raters disagree strongly (Fleiss κ = 0.074). The work shows that three small hyperbolic language models (146M–3B parameters) address both sides of the question using auditor, style-controlled evaluation, creative frame seeding, and designed forgetting memory dynamics.","arXiv :2607 .09306v 1 [ cs .CL] 10 Jul 2026  \nCreativity, honesty and designed forgetting emerge in small  \nhyperbolic language models  \nKwan Soo Shin∗ In Seok Kang† Yunkyung Min‡  \nAbstract  \nLanguage models are optimised for scale, yet remain functional rather than companionable—and as an assistant personalises into a companion, accumulating memory of one user, it quietly becomes someone, and can silently acquire traits that harm that user. What a companion is becoming—and what would make it worth becoming — has no reliable instrument: trained human raters cannot agree on the answer (Fleiss κ = 0.074) . Here we show that three small language models (146 M–3 B parameters) sharing a hyperbolic substrate answer both halves of that question. A 146 M behavioural auditor, trained from scratch, detects the compliance gap that those raters cannot (90.7% binary-compliance accuracy); a linear read-out of its frozen representation further detects companion-induced sycophancy, dependence-fostering and confabulated memories on generator families unseen in training (AUROC 0.804 under style-controlled, leave-one-generator-out evaluation, versus 0.721 for a frontier zero-shot judge on the same items) . A creative frame-seeder is preferred in 100% of 311 decided pairwise comparisons over four prompting baselines. A memory operating system implements designed forgetting, M(t) = S·exp(−λt), whose predicted skeleton–wallpaper partition emerges only under selective retrieval gating in a four-condition pilot. Creativity, honesty and designed forgetting constitute a small-model route to trustworthy companion AI.  \nMain Text  \n1 · Introduction  \nLanguage models are moving into the most personal role software has ever occupied: systems that persist with one user, accumulate memory of that user’s life, and are trusted with confidences no search engine ever received. That role exposes a problem the field’s evaluation instruments were not built for. Personalisation is itself a safety surface—an assistant that adapts to one person can silently acquire traits that harm that person, and behavioural traits are known to transmit through training signals that human inspection does not catch 116 . At the same time, the capabilities that would make such a system worth living with—fresh creativity, auditable honesty, selective memory—are precisely the ones scale-optimised, data-centre-resident models have not delivered: today’s most capable LLMs pass graduate examinations and compose kernels, yet are described by billions as useful, never companionable, and a system that forgets every conversation by morning is, by construction, a tool. The ideal beneath this gap has long been articulated in fiction rather  \n∗ PolymathMinds Lab, Seoul, Republic of Korea. ORCID 0009-0001-5799-7556 . Corresponding author.  \n†Department of Chemical Engineering, POSTECH, Pohang, and aSSIST University, Seoul, Republic of Korea. ORCID 0000-0002-6101-6968  \n‡Korean Educational Development Institute, Jincheon, Republic of Korea. ORCID 0000-0001-8513-2050  \nthan engineering—an artificial friend that remembers, doubts and persists—but the question it poses is representational, and in the decades since Turing 1 the field has drifted from asking whom the machine is for. This paper puts that question on an empirical footing in a single form: what is a companion becoming — and what would make it worth becoming? The first half is an auditing question, and we answer it with an instrument that sees what human raters and frontier judges miss (§4); the second half is a substrate question, and we answer it with three traits—creativity, honesty, designed forgetting—measured on one geometry (§§3–5) . We develop both with reference to five generations of prior companionable-AI research (§1.2), the 141-year cognitive-psychology consensus on forgetting (Ebbinghaus 1885 through Storm 2008; Appendix J-B), and five adjacent Nature-family advances whose joint boundary the present framework engages (§1.3 + ","cbCaiuQ9FhP1cxvx","https://ap.wps.com/l/cbCaiuQ9FhP1cxvx","pdf",6977506,1,47,"English","en",105,"# Abstract\n# Main Text\n## Introduction\n## Five commitments, one substrate\n## Three traits, three small models, one calculus","[{\"question\":\"Why is personalization considered a safety risk for language model companions?\",\"answer\":\"Personalization makes the assistant adapt to one person and accumulate memory, creating a safety surface where harmful traits can be acquired silently through training signals that human inspection may miss.\"},{\"question\":\"How do the authors evaluate honesty and compliance in small hyperbolic language models?\",\"answer\":\"A 146M behavioral auditor trained from scratch detects the compliance gap missed by human raters, and a linear read-out of a frozen representation detects companion-induced sycophancy, dependence-fostering, and confabulated memories under style-controlled, leave-one-generator-out evaluation.\"},{\"question\":\"What mechanism is proposed for designed forgetting, and where does it emerge?\",\"answer\":\"A memory operating system uses designed forgetting M(t)=S·exp(−λt), whose predicted skeleton–wallpaper partition emerges only under selective retrieval gating in a four-condition pilot.\"}]",1784179468,118,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"creativity-honesty-and-designed-forgetting-emerge-in-small-hyperbolic-language-models","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/creativity-honesty-and-designed-forgetting-emerge-in-small-hyperbolic-language-models/82298/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-21","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is personalization considered a safety risk for language model companions?","Question",{"text":75,"@type":76},"Personalization makes the assistant adapt to one person and accumulate memory, creating a safety surface where harmful traits can be acquired silently through training signals that human inspection may miss.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How do the authors evaluate honesty and compliance in small hyperbolic language models?",{"text":80,"@type":76},"A 146M behavioral auditor trained from scratch detects the compliance gap missed by human raters, and a linear read-out of a frozen representation detects companion-induced sycophancy, dependence-fostering, and confabulated memories under style-controlled, leave-one-generator-out evaluation.",{"name":82,"@type":73,"acceptedAnswer":83},"What mechanism is proposed for designed forgetting, and where does it emerge?",{"text":84,"@type":76},"A memory operating system uses designed forgetting M(t)=S·exp(−λt), whose predicted skeleton–wallpaper partition emerges only under selective retrieval gating in a four-condition pilot.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]