[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82367-en":3,"doc-seo-82367-105":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},82367,687197207919,"Theodora","https://ap-avatar.wpscdn.com/avatar/a000253d6f5f7c60be?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779446848396160552",8,"Research & Report","Shared Selective Persistent Memory for Agentic LLM Systems","Agentic LLM systems that write code via multi-turn tool use suffer from a core context loss: each session starts empty, discarding domain constraints, data schemas, tool configurations, and productive prompt patterns. The work proposes shared selective persistent memory that retains four reusable context categories—task specifications, data schemas, tool configurations, and output constraints—while discarding session-specific reasoning traces. Deployed in a collaborative Git-versioned workspace, it enables transferable memory with role-based access and supports zero-token data refresh for recurring updates. Across three enterprise scenarios, selective memory achieves major task completion gains (96% vs 79% without memory).","Shared Selective Persistent Memory for Agentic LLM Systems  \narXiv :2607 .09493v 1 [ cs .AI] 10 Jul 2026  \nSanjana Pedada  \nApple Inc. [s_pedada@apple.com](s_pedada@apple.com)  \nAditya Dhavala  \nApple Inc. [a_dhavala@apple.com](a_dhavala@apple.com)  \nNeelraj Patil  \nApple Inc. [neelraj_patil@apple.com](neelraj_patil@apple.com)  \nAbstract  \ninvocation token cost by 97 × versus raw data injection. A replication on four public datasets  \nAgentic LLM systems that generate code through multi-turn tool use face a fundamental context problem: each session starts from zero, discarding the configuration choices, domain constraints, data schemas, and tooluse patterns that made previous sessions productive. Naively persisting entire conversation histories is both token-inefficient and counterproductive—irrelevant context degrades generation quality. We introduce shared selective persistent memory, a memory architecture for agentic systems that identifies and retains four categories of reusable context—task specifications, data schemas, tool configurations, and output constraints—while discarding session-specific reasoning traces. Crucially, this memory is shared: workspaces encapsulating selective memory can be transferred across users with role-based access control, enabling collaborative reuse of accumulated context without redundant specification. We implement this architecture ina deployed collaborative workspace platform where LLM agents produce, edit, and maintain git-versioned artifacts—including interactive dashboards, structured reports, and datadriven documents—from heterogeneous data sources accessed via multiple connector types (CSV upload, SQL, REST APIs, and MCP servers) . Git-backed versioning with draft isolation enables users to explore modifications risk-free and restore to any prior state without re-invoking the model. A complementary zero-token data refresh mechanism decouples generated programs from runtime data, enabling artifact reuse without re-invocation. Across three enterprise deployment scenarios, shared selective persistent memory achieves 96% task completion (vs. 79% without memory and 71% with full history) . A complementary zero-token data refresh mechanism eliminates LLM re-invocation entirely for recurring data updates (14× task time reduction), while summary-driven generation reduces per-  \nconfirms generalizability, with zero-token refresh succeeding in 12/12 trials. Notably, naive full-history persistence actively degrades task completion by biasing the agent with stale reasoning traces, while selective memory outperforms both extremes.  \n1 Introduction  \nAgentic LLM systems—those that autonomously invoke tools, write files, and execute code to accomplish user goals—have demonstrated impressive capabilities in code generation (GitHub, 2021 ; Anthropic, 2025b), data analysis (OpenAI, 2023b), and multi-step reasoning (Wu et al., 2023) . Yet a fundamental limitation persists across all current agentic frameworks: sessions are stateless. When a conversation ends, the accumulated context—domain constraints, data schemas, tool configurations, prompt refinements, and output format preferences—is discarded entirely.  \nThis is wasteful. In enterprise workflows, users repeatedly perform structurally similar tasks: generating weekly dashboards from updated data exports, producing reports with consistent formatting, querying the same internal tools with the same authentication patterns, drafting documents from evolving data sources, and maintaining versioned artifacts across teams. Each session forces the user to re-specify what the system should already know. The problem is not that LLMs lack memory, but that existing systems provide no mechanism to selectively persist the context that matters while discarding the session-specific reasoning that does not.  \nThe naive solution—persisting entire conversation histories—is counterproductive. Prior work on long-context LLMs (Liu et al., 2024) demonstrates that irrele","cbCaispQboGi136r","https://ap.wps.com/l/cbCaispQboGi136r","pdf",220033,1,11,"English","en",105,"# Abstract\n# Introduction\n## Stateless session context problem\n## Drawbacks of full-history persistence\n## Proposed shared selective persistent memory\n## Reusable context categories\n## Discarded session-specific traces","[{\"question\":\"What fundamental limitation do agentic LLM systems face in current frameworks?\",\"answer\":\"Sessions are effectively stateless, so when one session ends the accumulated context—constraints, schemas, tool configurations, and prompt refinements—is discarded entirely.\"},{\"question\":\"Why is persisting the entire conversation history harmful for agentic systems?\",\"answer\":\"Irrelevant context can degrade generation, and prior tool-use traces bias the agent toward previously explored solution paths rather than the new task.\"},{\"question\":\"What does shared selective persistent memory retain and what does it discard?\",\"answer\":\"It retains four reusable context categories: task specifications, data schemas, tool configurations, and output constraints, while discarding session-specific reasoning traces, tool invocation logs, intermediate file states, and error-recovery paths.\"}]",1784179958,28,{"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},"shared-selective-persistent-memory-for-agentic-llm-systems","",{"@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/shared-selective-persistent-memory-for-agentic-llm-systems/82367/",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","2026-07-17","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 fundamental limitation do agentic LLM systems face in current frameworks?","Question",{"text":75,"@type":76},"Sessions are effectively stateless, so when one session ends the accumulated context—constraints, schemas, tool configurations, and prompt refinements—is discarded entirely.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Why is persisting the entire conversation history harmful for agentic systems?",{"text":80,"@type":76},"Irrelevant context can degrade generation, and prior tool-use traces bias the agent toward previously explored solution paths rather than the new task.",{"name":82,"@type":73,"acceptedAnswer":83},"What does shared selective persistent memory retain and what does it discard?",{"text":84,"@type":76},"It retains four reusable context categories: task specifications, data schemas, tool configurations, and output constraints, while discarding session-specific reasoning traces, tool invocation logs, intermediate file states, and error-recovery paths.","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"]