{"id":"https://openalex.org/W7162686460","doi":"https://doi.org/10.48550/arxiv.2605.27744","title":"A Policy-Driven Runtime Layer for Agentic LLM Serving","display_name":"A Policy-Driven Runtime Layer for Agentic LLM Serving","publication_year":2026,"publication_date":"2026-05-26","ids":{"openalex":"https://openalex.org/W7162686460","doi":"https://doi.org/10.48550/arxiv.2605.27744"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.27744","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.27744","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.27744","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137304093","display_name":"Rui Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Rui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103173191","display_name":"Chae-Eun Kim","orcid":"https://orcid.org/0009-0007-7929-4409"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Chaeeun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5113952097","display_name":"Liting Hu","orcid":"https://orcid.org/0000-0001-8512-5176"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Liting","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.19089999794960022,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.19089999794960022,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12203","display_name":"Mobile Agent-Based Network Management","score":0.1535000056028366,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.09709999710321426,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/abstraction","display_name":"Abstraction","score":0.6559000015258789},{"id":"https://openalex.org/keywords/prefix","display_name":"Prefix","score":0.6111000180244446},{"id":"https://openalex.org/keywords/event","display_name":"Event (particle physics)","score":0.5925999879837036},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.5871999859809875},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.5253999829292297},{"id":"https://openalex.org/keywords/scheme","display_name":"Scheme (mathematics)","score":0.4754999876022339},{"id":"https://openalex.org/keywords/point","display_name":"Point (geometry)","score":0.47519999742507935},{"id":"https://openalex.org/keywords/state","display_name":"State (computer science)","score":0.412200003862381},{"id":"https://openalex.org/keywords/identity","display_name":"Identity (music)","score":0.3758000135421753}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7487000226974487},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.6559000015258789},{"id":"https://openalex.org/C141603448","wikidata":"https://www.wikidata.org/wiki/Q134830","display_name":"Prefix","level":2,"score":0.6111000180244446},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.5925999879837036},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.5871999859809875},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.5253999829292297},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.4754999876022339},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.47519999742507935},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.412200003862381},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.39309999346733093},{"id":"https://openalex.org/C2778355321","wikidata":"https://www.wikidata.org/wiki/Q17079427","display_name":"Identity (music)","level":2,"score":0.3758000135421753},{"id":"https://openalex.org/C9395851","wikidata":"https://www.wikidata.org/wiki/Q177929","display_name":"Stack (abstract data type)","level":2,"score":0.36910000443458557},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.36910000443458557},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.3513999879360199},{"id":"https://openalex.org/C174839445","wikidata":"https://www.wikidata.org/wiki/Q1134386","display_name":"Lock (firearm)","level":2,"score":0.3319000005722046},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.3310999870300293},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.30660000443458557},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.30640000104904175},{"id":"https://openalex.org/C136815107","wikidata":"https://www.wikidata.org/wiki/Q154242","display_name":"Nothing","level":2,"score":0.30390000343322754},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.3012999892234802},{"id":"https://openalex.org/C189950617","wikidata":"https://www.wikidata.org/wiki/Q937228","display_name":"Property (philosophy)","level":2,"score":0.29739999771118164},{"id":"https://openalex.org/C147358964","wikidata":"https://www.wikidata.org/wiki/Q1200992","display_name":"Abstraction layer","level":3,"score":0.2736000120639801},{"id":"https://openalex.org/C96147967","wikidata":"https://www.wikidata.org/wiki/Q190686","display_name":"Subroutine","level":2,"score":0.2678999900817871},{"id":"https://openalex.org/C189783530","wikidata":"https://www.wikidata.org/wiki/Q352090","display_name":"CPU cache","level":3,"score":0.26669999957084656},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.2606000006198883},{"id":"https://openalex.org/C144240696","wikidata":"https://www.wikidata.org/wiki/Q367204","display_name":"Address space","level":2,"score":0.26030001044273376},{"id":"https://openalex.org/C2781140086","wikidata":"https://www.wikidata.org/wiki/Q557945","display_name":"Confusion","level":2,"score":0.257999986410141},{"id":"https://openalex.org/C2778348673","wikidata":"https://www.wikidata.org/wiki/Q739302","display_name":"Production (economics)","level":2,"score":0.2506999969482422}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.27744","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.27744","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.27744","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.27744","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multi-agent":[0],"LLM":[1],"systems":[2],"have":[3],"become":[4],"the":[5,10,36,73,76,91,117,120,139,148,152,157,160,174],"dominant":[6],"production":[7],"workload,":[8],"but":[9,30,43],"serving":[11,37,219],"stack":[12],"was":[13],"not":[14],"built":[15],"for":[16,183],"them.":[17],"The":[18],"agent":[19,23,113,136,176],"framework":[20,118],"above":[21],"knows":[22,44],"identities,":[24],"role,":[25],"schemas,":[26],"and":[27,68,79,119,150,180,186,210],"dispatch":[28],"structure":[29],"never":[31],"sees":[32,40],"an":[33,101,112,217],"engine-level":[34],"event;":[35],"engine":[38],"below":[39],"every":[41],"event":[42],"nothing":[45],"about":[46],"agents.":[47],"A":[48],"surprising":[49],"number":[50],"of":[51],"cross-cutting":[52],"policies":[53,146],"depend":[54],"on":[55,156,191],"both:":[56],"prefix":[57],"caching,":[58],"batch":[59],"shaping,":[60],"speculative":[61],"execution,":[62],"fairness,":[63],"tool-result":[64],"memoization,":[65],"safety":[66],"enforcement,":[67],"more.":[69],"Each":[70],"lives":[71],"in":[72,154],"seam":[74,96],"between":[75,116],"two":[77],"layers":[78],"is":[80,97],"currently":[81],"solved":[82],"by":[83,100],"a":[84,109],"one-off":[85],"patch":[86],"into":[87,129],"one":[88,158],"neighbor":[89],"or":[90],"other.":[92],"We":[93,142],"argue":[94],"this":[95],"best":[98],"addressed":[99],"architectural":[102],"change":[103],"rather":[104],"than":[105],"point":[106],"fixes:":[107],"insert":[108],"third":[110],"tier,":[111],"runtime":[114],"layer,":[115],"engine,":[121],"exposing":[122],"four":[123],"primitives":[124],"(observe,":[125],"score,":[126],"predict,":[127],"act)":[128],"which":[130,172],"any":[131],"agent-aware":[132],"policy":[133],"plugs,":[134],"with":[135,159],"identity":[137],"as":[138,170],"shared":[140],"coordinate.":[141],"map":[143],"nine":[144],"concrete":[145],"onto":[147],"layer":[149],"validate":[151],"abstraction":[153],"depth":[155],"largest":[161],"immediate":[162],"serving-cost":[163],"lever:":[164],"KV":[165],"caching":[166],"across":[167],"sessions,":[168],"instantiated":[169],"CacheSage,":[171],"learns":[173],"per-workload":[175],"transition":[177],"matrix":[178],"online":[179],"uses":[181],"it":[182],"survival-based":[184],"eviction":[185],"between-step":[187],"prefetch.":[188],"Preliminary":[189],"results":[190],"five":[192],"real":[193],"multi-agent":[194],"workloads":[195],"show":[196],"+13":[197],"to":[198,205,212],"+37":[199],"pp":[200],"cache":[201],"hit-rate":[202],"lift,":[203],"12%":[204],"29%":[206],"lower":[207],"mean":[208],"TTFT,":[209],"6%":[211],"14%":[213],"higher":[214],"throughput":[215],"over":[216],"unmodified":[218],"stack.":[220]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-29T00:00:00"}
