[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85193-en":3,"doc-seo-85193-105":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},85193,962075114101,"Seraphina","https://ap-avatar.wpscdn.com/avatar/e000253a75eb197efd?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780044092746381165",6,"Technology","TGMS: An Agent-Native Bi-Temporal Graph Management System","TGMS is an agent-native bi-temporal property graph management system for temporal graph questions that require correct handling of valid time, transaction time, identifiers, and arithmetic. It exposes thirteen verified temporal operators as typed, deterministic, bounded, cost-guarded tools. An LLM planner proposes operator calls, while the system performs all computation and checks numeric, entity, ordering, and pattern claims against a content-addressed execution trace. Separating valid time from transaction time enables belief-state queries, including correction-sensitive questions.","arXiv :2607 . 10265v1 [ cs .DB] 11 Jul 2026  \nTGMS: An Agent-Native Bi-Temporal Graph Management System  \nVerified Temporal Operators and Trace-Grounded Answer Checking  \nXiaofei Zhang  \nUniversity of Memphis  \n[xiaofei. zhang@memphis. edu](xiaofei. zhang@memphis. edu)  \nJuly 2026 · System Description Preprint (v1) · [https://github. com/zxf-work/tgms](https://github. com/zxf-work/tgms) ∗  \nAbstract  \nTemporal graph questions require reliable handling of time, identifiers, and arithmetic. Large language model (LLM) agents often fail on these tasks, especially when a graph records both ordinary evolution and later corrections. We present TGMS, a bi-temporal property graph management system that exposes thirteen verified temporal operators as agent tools. Each operator is typed, deterministic, bounded, cost-guarded, and bi-temporal by default. The LLM plans operator calls and writes the final response, while the system performs all graph computation. Numeric, entity, ordering, and pattern claims are checked against the content-addressed execution trace.  \nTGMS separates valid time from transaction time. It can therefore answer belief-state questions such as “as of transaction time T, what did the system believe?” Standard latest-state snapshots and retrieval pipelines do not preserve enough information to answer such questions. On a development benchmark built from a real communication network, TGMS with a 14B opensource model reaches 0.409 exact match. Vector-RAG, static-graph RAG, and text-to-Cypher reach 0.045–0.182 under the same serving setup. TGMS reaches 0.67 exact match on correction probes, while the three 14B baselines score zero. The claim verifier detects all 500 injected count and entity errors with no false positives on the clean answers. Two implementation findings were especially important. First, operator output contracts prevent plans from referring to fields that do not exist. Second, verification must track whether the cited evidence is complete, because correct arithmetic over a truncated result is still misleading. The code, benchmark, and trace viewer are open source under Apache-2 .0.  \n1 Introduction  \nTemporal graphs arise in communication logs, transaction records, and evolving knowledge bases. They create two difficulties for LLM agents that are less visible in static text collections.  \nThe first difficulty is temporal composition. Consider the question: “Among accounts reachable from X in February, how many cyclic triangles closed within one day?” Answering it requires timerespecting reachability [28], followed by δ-temporal motif counting [18] . A retrieval pipeline that serializes edges into text may omit the exact structure needed by the second step.  \nThe second difficulty is belief revision. A graph may change because the world changed, or because an earlier record was wrong. These cases are different. Answering “what did we believe  \n∗ This preprint accompanies TGMS v0 .1.0 and reports results on the development split. A revised version will report the frozen-test campaign, whose acceptance thresholds are recorded in the repository’s dated decision log.  \non March 1?” requires both valid time and transaction time [25] . A latest-state snapshot cannot reconstruct this distinction after a correction has been applied.  \nLLMs introduce a separate set of risks. They may perform arithmetic incorrectly, invent identifiers, or report values that do not appear in the evidence. TGMS addresses these risks through system design rather than prompting. Identifiers must come from the task input or from an entity-resolution operator. Arithmetic must use a compute operator. Each final claim must cite a content-addressed result, and a verifier checks the cited evidence.  \nThis paper makes four contributions:  \n• We develop a bi-temporal property graph substrate with explicit operations for assertion, retraction, and correction. An append-only event log makes updates replayable across storage backends.  \n• We de","cbCaigMwumK53d08","https://ap.wps.com/l/cbCaigMwumK53d08","pdf",302681,1,12,"English","en",105,"# Abstract\n# Introduction\n## Temporal composition\n## Belief revision and temporal semantics\n# Bi-temporal substrate\n## Versions and time intervals\n## Write operations and event log","[{\"question\":\"What problem does TGMS address in temporal graph QA?\",\"answer\":\"TGMS targets unreliable handling of time (valid vs transaction), identifiers, and arithmetic in temporal graph questions, especially when graphs include both normal evolution and later corrections.\"},{\"question\":\"How does TGMS ensure the LLM’s final answers are trustworthy?\",\"answer\":\"TGMS performs all graph computation and verifies final claims against a content-addressed execution trace, checking numeric, entity, ordering, and pattern claims grounded in evidence.\"},{\"question\":\"What is the role of separating valid time and transaction time?\",\"answer\":\"Separating valid time from transaction time enables belief-state questions, such as what the system believed at a specific transaction time, which latest-state snapshots cannot reconstruct after corrections.\"}]",1784201653,30,{"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},"tgms-an-agent-native-bi-temporal-graph-management-system","",{"@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/technology/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/tgms-an-agent-native-bi-temporal-graph-management-system/85193/",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 problem does TGMS address in temporal graph QA?","Question",{"text":75,"@type":76},"TGMS targets unreliable handling of time (valid vs transaction), identifiers, and arithmetic in temporal graph questions, especially when graphs include both normal evolution and later corrections.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does TGMS ensure the LLM’s final answers are trustworthy?",{"text":80,"@type":76},"TGMS performs all graph computation and verifies final claims against a content-addressed execution trace, checking numeric, entity, ordering, and pattern claims grounded in evidence.",{"name":82,"@type":73,"acceptedAnswer":83},"What is the role of separating valid time and transaction time?",{"text":84,"@type":76},"Separating valid time from transaction time enables belief-state questions, such as what the system believed at a specific transaction time, which latest-state snapshots cannot reconstruct after corrections.","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},[93,97,101,105,110,113,118,122,127,130,134],{"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":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":111,"slug":112},50,"technology",{"id":114,"doc_module":4,"doc_module_name":45,"category_name":115,"show_sort_weight":116,"slug":117},7,"Healthcare",40,"healthcare",{"id":119,"doc_module":4,"doc_module_name":45,"category_name":120,"show_sort_weight":28,"slug":121},8,"Research & Report","research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]