[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85841-en":3,"doc-seo-85841-105":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},85841,8796095461610,"Oliver","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","GRATE Temporal Extensions for Inductive KG Foundation Models via Gated Rotary Attention","Knowledge graph foundation models such as ULTRA and TRIX provide strong inductive transfer by learning relation-graph representations that generalize to unseen entities and relations. Extending this capability to temporal knowledge graphs is difficult because prior temporal models bind parameters to dataset-specific entities, relations, and timestamps. GRATE proposes a parameter-free gated rotary attention mechanism that encodes time using relative time gaps, integrates into NBFNet-style models, and improves cross-dataset temporal transfer on newly built benchmark suites.","GRATE: Temporal Extensions for Inductive KG Foundation Models  \nvia Gated Rotary Attention  \nJiaxin Pan 1 Osama Mohammed 1 Daniel Hernndez 1 Steffen Staab 1 2  \narXiv :2607 . 10 197v 1 [ cs .AI] 11 Jul 2026  \nAbstract  \nKnowledge graph foundation models such as ULTRA and TRIX achieve strong inductive transfer by learning relation-graph representations that generalise to unseen entities and relations. Extending this transferability to temporal knowledge graphs (TKGs) remains challenging: existing temporal models tie their parameters to datasetspecific entities, relations, or timestamps and are not designed to transfer to TKGs with disjoint vocabularies. We propose GRATE (Gated Rotary Attention for Temporal Encoding), an entity-side message function that adds no learnable parameters and encodes time through relative time differences by rotating each edge message according to its time gap to the query and applying a query-conditioned gate to select temporally relevant signals. GRATE integrates into NBFNetstyle KG foundation models while preserving structural transferability. Existing TKG benchmarks evaluate within shared train/test vocabularies and cannot directly test cross-dataset temporal transfer; we therefore construct GDELTINDT and WIKIINDT, inductive transfer benchmark suites with disjoint entities, relations, and timestamps spanning both interpolation and extrapolation. Across these benchmarks and held-out forecasting datasets, a single jointly pretrained GRATE checkpoint improves over the static base model in most settings.  \n1. Introduction  \nRecent knowledge graph (KG) foundation models (Galkinet al., 2024 ; Zhang et al., 2025 ; Du et al., 2026) learn  \nAccepted at the ICML 2026 Workshop on Graph Foundation Models: A New Era for Graph Machine Learning. 1Analytic Computing, Institute for Artificial Intelligence, University of Stuttgart, Stuttgart, Germany 2University of Southampton, Southampton, United Kingdom. Correspondence to: Jiaxin Pan \u003C[jiaxin.pan@ki.uni-stuttgart.de](jiaxin.pan@ki.uni-stuttgart.de) >.  \nPreprint. July 14, 2026.  \nvocabulary-agnostic representations that transfer to graphs with unseen entities and relations, supporting link prediction queries such as “who did Obama engage in negotiation with?” Many facts are also inherently temporal: temporal knowledge graphs (TKGs) attach a timestamp to each fact, so the same query is only meaningful when paired with a year (“. . . in 2010”) . Existing TKG models (Zhang et al., 2022 ; Li et al., 2021 ; Sun et al., 2021) learn datasetspecific embeddings for entities, relations, and timestampsand must be retrained whenever new ones appear, making them unsuitable for settings such as emerging political actors, newly discovered scientists, or rapidly evolving event streams, where new vocabulary arrives continuously.  \nA natural alternative is to apply an inductive KG foundation model directly to a TKG. However, temporal reasoning requires similarity judgements that are not purely structural. Inductive KG foundation models transfer by comparing new entities and relations through their graph contexts, but in a TKG the usefulness of a supporting fact also depends on its time and its relevance to the query. For a query such as “who did the US president meet in 2024?”, a meeting from 2024 is usually more similar to the query context than one from 1974, even if both have similar graph structure. Likewise, among same-day events, a meeting with a head of state is more relevant to a diplomatic query than a meeting with a sports figure. Thus, inductive temporal reasoning requires the model to judge supporting facts not only by structural similarity, but also by relative temporal displacement and query-conditioned relevance. Existing inductive KG foundation models do not provide this: they define similarity only through static graph structure, treating facts from all timestamps as equally aligned with the query and aggregating all supporting facts indiscriminately.  \nTherefore,","cbCaicSK0n441J4e","https://ap.wps.com/l/cbCaicSK0n441J4e","pdf",911917,3,1,17,"English","en",105,"# Abstract\n# Introduction\n# Related Work\n## Temporal Knowledge Graph Embedding\n## Transductive Interpolation (TKG Completion)","[{\"question\":\"What problem does GRATE address in temporal knowledge graphs?\",\"answer\":\"Existing temporal models often require retraining when entities, relations, or timestamps change, and existing inductive KG foundation models ignore temporal alignment. GRATE targets this by adding temporal reasoning that transfers to disjoint temporal vocabularies.\"},{\"question\":\"How does GRATE encode time without adding learnable parameters?\",\"answer\":\"GRATE rotates each edge message according to its relative time gap to the query and applies a query-conditioned gate to weight the temporally relevant evidence.\"},{\"question\":\"How do the authors evaluate GRATE for inductive temporal transfer?\",\"answer\":\"They build inductive transfer benchmark suites with disjoint entities, relations, and timestamps and test GRATE on those benchmarks plus held-out forecasting datasets, comparing against static base models.\"}]",1784206642,43,{"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},"grate-temporal-extensions-for-inductive-kg-foundation-models-via-gated-rotary-attention","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":20},"https://docshare.wps.com/document/research-report/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/grate-temporal-extensions-for-inductive-kg-foundation-models-via-gated-rotary-attention/85841/",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","2026-07-25","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 GRATE address in temporal knowledge graphs?","Question",{"text":75,"@type":76},"Existing temporal models often require retraining when entities, relations, or timestamps change, and existing inductive KG foundation models ignore temporal alignment. GRATE targets this by adding temporal reasoning that transfers to disjoint temporal vocabularies.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does GRATE encode time without adding learnable parameters?",{"text":80,"@type":76},"GRATE rotates each edge message according to its relative time gap to the query and applies a query-conditioned gate to weight the temporally relevant evidence.",{"name":82,"@type":73,"acceptedAnswer":83},"How do the authors evaluate GRATE for inductive temporal transfer?",{"text":84,"@type":76},"They build inductive transfer benchmark suites with disjoint entities, relations, and timestamps and test GRATE on those benchmarks plus held-out forecasting datasets, comparing against static base models.","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},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"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":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]