[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-81756-en":3,"doc-seo-81756-105":30,"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":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},81756,4398048949847,"Eliana","https://ap-avatar.wpscdn.com/avatar/400002536579ef2da7f?_k=1778318612642679267",8,"Research & Report","From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents","How do two agents invent a shared language from scratch? In a Lewis signaling game, a sender and receiver coordinate on a code using only their interaction history. This study compares five memory architectures across different channel configurations with LLM agents and finds that memory architecture outweighs channel capacity. Persistent private notebooks enable more reliable coordination (0.867 ± 0.023 at capacity 25) by externalizing conventions and avoiding high-capacity collapse. Stateless agents peak then degrade as learned vocabulary exceeds rolling context tracking. An information-bottleneck argument predicts an optimal capacity equal to object count, but the bottleneck becomes a fragility point.","From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents  \nYashar Talebirad 1 , Eden Redman2 , Ali Parsaee 1 , and Osmar R. Za1  \n1Alberta Machine Intelligence Institute, University of Alberta, Edmonton, Canada  \n2Network for Applied Technology, Edmonton, Canada  \n[talebira@ualberta.ca](talebira@ualberta.ca), [eden@nat.ltd](eden@nat.ltd)  \narXiv :2607 .00233v 1 [ cs .AI] 30 Jun 2026  \nAbstract  \nHow do two agents invent a shared language from scratch? In a Lewis signaling game, a sender and receiver must coordinate on a code using only their interaction history. We study five memory architectures across varying channel configurations with LLM agents and find that memory architecture matters more than channel capacity. Agents with a persistent private notebook benefit from surplus channel capacity and avoid the high-capacity collapse seen in stateless agents, achieving the most reliable coordination (0 .867 ± 0.023 at capacity = 25) . Stateless agents peak at moderate capacity and then degrade as the vocabulary grows beyond what a rolling context window can track. The notebook externalizes learned conventions, freeing agents from having to rederive codes each round. An information bottleneck-inspired argument predicts an optimal capacity equal to the number of objects. Instead, the bottleneck (capacity = 8) proves to bea fragility point, and surplus capacity is generally better. We show that channel capacity alone cannot predict coordination; memory architecture determines whether agents turn interaction history into stable conventions, and both dimensions are needed to understand how signals become language.  \nIntroduction  \nThe Lewis signaling game (Lewis, 1969) is a minimal model of communication emergence: a sender observes a designated target among a set of candidates and transmits a constrained signal, and a receiver sees the same candidates and the signal, then identifies the target. No pre-negotiated meanings exist, and agents converge through repeated coordination alone. Consistent success in this coordination task means that a new language has been invented by the agents. Because the objects are described by familiar features and the agents are pretrained models with semantic priors, this language is a mapping from an arbitrary signal space onto an already structured meaning space.  \n©2026 Yashar Talebirad, Eden Redman, Ali Parsaee, and Osmar R. Za . Published under a Creative Commons Attribution 4.0 International (CC BY 4 .0) license.  \nThis is the authors’ version of a paper accepted to ALIFE 2026, with minor aesthetic changes from the version of record, which appears in the ALIFE 2026 conference proceedings.  \nLarge language models introduce a different kind of agent. Unlike gradient-trained agents, LLMs provide a general-purpose reasoner that can be placed in a wide range of simulated environments without retraining, bringing linguistic and inferential priors to each task. LLMs also have the ability to adapt through in-context learning (Brown et al., 2020): they can reason over a history of prior interactions to refine strategy on each new call. This shifts attention from model architecture to memory architecture: what information each agent retains across rounds and in what form. The scratchpad (Nye et al., 2021) and chain-of-thought (Wei et al., 2022) literatures show that the structure of intermediate representations shapes what LLMs can compute. Ina signaling game, how an agent stores what it has learned determines the language it can invent. Classical emergent communication research uses this framework with gradienttrained neural agents, finding that compositional protocols, those in which signal structure mirrors object structure, emerge under the right pressures (Lazaridou et al., 2017) . Information-theoretic bottleneck arguments (Tishby et al., 1999) motivate paying special attention when the channel is scarce relative to the number of referents. Resnick et al.(2020) stud","cbCaim80RimAEpG5","https://ap.wps.com/l/cbCaim80RimAEpG5","pdf",1347574,2,1,9,"English","en",105,"# Abstract\n# Introduction\n# Background\n## Lewis signaling games\n## LLMs and memory architecture","[{\"question\":\"What question does the paper address about language emergence in LLM agents?\",\"answer\":\"It investigates how agents invent a shared language from scratch and whether memory architecture or channel capacity is the key driver of successful coordination.\"},{\"question\":\"How is the experiment framed?\",\"answer\":\"The study uses a Lewis signaling game where agents must coordinate on a code based only on their interaction history, without any pre-negotiated meanings.\"},{\"question\":\"What is the main finding about memory architecture vs. channel capacity?\",\"answer\":\"Memory architecture matters more than channel capacity: persistent private notebooks enable stable, reliable coordination, while stateless agents degrade as the code space grows beyond what rolling context can track.\"}]",1784175851,23,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"from-signals-to-structure-how-memory-architecture-drives-language-emergence-in-llm-agents","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":20},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/from-signals-to-structure-how-memory-architecture-drives-language-emergence-in-llm-agents/81756/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","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},"What question does the paper address about language emergence in LLM agents?","Question",{"text":75,"@type":76},"It investigates how agents invent a shared language from scratch and whether memory architecture or channel capacity is the key driver of successful coordination.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is the experiment framed?",{"text":80,"@type":76},"The study uses a Lewis signaling game where agents must coordinate on a code based only on their interaction history, without any pre-negotiated meanings.",{"name":82,"@type":73,"acceptedAnswer":83},"What is the main finding about memory architecture vs. channel capacity?",{"text":84,"@type":76},"Memory architecture matters more than channel capacity: persistent private notebooks enable stable, reliable coordination, while stateless agents degrade as the code space grows beyond what rolling context can track.","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":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]