{"id":"https://openalex.org/W7159576906","doi":"https://doi.org/10.48550/arxiv.2604.27536","title":"Belief-Guided Inference Control for Large Language Model Services via Verifiable Observations","display_name":"Belief-Guided Inference Control for Large Language Model Services via Verifiable Observations","publication_year":2026,"publication_date":"2026-04-30","ids":{"openalex":"https://openalex.org/W7159576906","doi":"https://doi.org/10.48550/arxiv.2604.27536"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.27536","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.27536","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":null,"license_id":null,"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.2604.27536","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134977737","display_name":"Wenhao Yuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Yuan, Wenhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134974447","display_name":"Chenchen Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Chenchen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134932764","display_name":"Jian Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Jian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134993810","display_name":"Jinfeng Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Jinfeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134970366","display_name":"Shuo Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Shuo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5123433413","display_name":"Edith Cheuk Han Ngai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ngai, Edith Cheuk Han","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5134977737"],"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.13199999928474426,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.13199999928474426,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T10028","display_name":"Topic Modeling","score":0.1316000074148178,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T13553","display_name":"Age of Information Optimization","score":0.046300001442432404,"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/inference","display_name":"Inference","score":0.7890999913215637},{"id":"https://openalex.org/keywords/observability","display_name":"Observability","score":0.7211999893188477},{"id":"https://openalex.org/keywords/partially-observable-markov-decision-process","display_name":"Partially observable Markov decision process","score":0.6581000089645386},{"id":"https://openalex.org/keywords/verifiable-secret-sharing","display_name":"Verifiable secret sharing","score":0.6166999936103821},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.4230000078678131},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.3833000063896179},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.3765999972820282},{"id":"https://openalex.org/keywords/markov-decision-process","display_name":"Markov decision process","score":0.37049999833106995},{"id":"https://openalex.org/keywords/state","display_name":"State (computer science)","score":0.3686000108718872}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7890999913215637},{"id":"https://openalex.org/C36299963","wikidata":"https://www.wikidata.org/wiki/Q1369844","display_name":"Observability","level":2,"score":0.7211999893188477},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7067999839782715},{"id":"https://openalex.org/C17098449","wikidata":"https://www.wikidata.org/wiki/Q176814","display_name":"Partially observable Markov decision process","level":4,"score":0.6581000089645386},{"id":"https://openalex.org/C85847156","wikidata":"https://www.wikidata.org/wiki/Q59015987","display_name":"Verifiable secret sharing","level":3,"score":0.6166999936103821},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.4230000078678131},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4032000005245209},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.3833000063896179},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.3765999972820282},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3752000033855438},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.37049999833106995},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.3686000108718872},{"id":"https://openalex.org/C28901747","wikidata":"https://www.wikidata.org/wiki/Q177571","display_name":"Decision theory","level":2,"score":0.3447999954223633},{"id":"https://openalex.org/C32848918","wikidata":"https://www.wikidata.org/wiki/Q845789","display_name":"Observable","level":2,"score":0.34380000829696655},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.34369999170303345},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.33980000019073486},{"id":"https://openalex.org/C163836022","wikidata":"https://www.wikidata.org/wiki/Q6771326","display_name":"Markov model","level":3,"score":0.33820000290870667},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.3370000123977661},{"id":"https://openalex.org/C98763669","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov chain","level":2,"score":0.3271999955177307},{"id":"https://openalex.org/C917703","wikidata":"https://www.wikidata.org/wiki/Q7239668","display_name":"Predictive inference","level":5,"score":0.304500013589859},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.302700012922287},{"id":"https://openalex.org/C84839998","wikidata":"https://www.wikidata.org/wiki/Q5249245","display_name":"Decision rule","level":2,"score":0.2653999924659729},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.26409998536109924},{"id":"https://openalex.org/C2777472644","wikidata":"https://www.wikidata.org/wiki/Q16968992","display_name":"Approximate inference","level":3,"score":0.2621000111103058},{"id":"https://openalex.org/C59594135","wikidata":"https://www.wikidata.org/wiki/Q5249242","display_name":"Decision model","level":2,"score":0.259799987077713},{"id":"https://openalex.org/C92991967","wikidata":"https://www.wikidata.org/wiki/Q7644329","display_name":"Supervisory control","level":3,"score":0.2565000057220459},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.25600001215934753},{"id":"https://openalex.org/C72434380","wikidata":"https://www.wikidata.org/wiki/Q230930","display_name":"State space","level":2,"score":0.25380000472068787},{"id":"https://openalex.org/C134261354","wikidata":"https://www.wikidata.org/wiki/Q938438","display_name":"Statistical inference","level":2,"score":0.251800000667572}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.27536","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.27536","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.27536","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.27536","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[{"score":0.40241602063179016,"display_name":"Partnerships for the goals","id":"https://metadata.un.org/sdg/17"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"In":[0,57],"black-box":[1,76],"large":[2],"language":[3],"model":[4],"(LLM)":[5],"services,":[6],"response":[7,42,55,120],"reliability":[8],"is":[9,43,123],"often":[10],"only":[11],"partially":[12],"observable":[13,86],"at":[14],"decision":[15,29,88],"time,":[16],"while":[17],"stronger":[18,155],"inference":[19,73,142,164],"pathways":[20],"incur":[21],"substantial":[22],"computational":[23],"cost,":[24],"inducing":[25],"a":[26,69,84,100,115,127,140],"budgeted":[27],"sequential":[28,95],"problem:":[30],"for":[31,64,71],"each":[32],"request,":[33],"the":[34,39,106,135],"system":[35],"should":[36,50],"decide":[37,131],"whether":[38,47,132],"default":[40,136],"low-cost":[41],"sufficiently":[44],"reliable":[45],"or":[46,138],"additional":[48],"computation":[49],"be":[51],"allocated":[52],"to":[53,90,130,133],"improve":[54],"quality.":[56],"this":[58],"paper,":[59],"we":[60],"propose":[61],"\\textbf{Ver}ifiable":[62],"\\textbf{O}bservations":[63],"Risk-aware":[65],"\\textbf{I}nference":[66],"\\textbf{C}ontrol":[67],"(\\textsc{Veroic}),":[68],"framework":[70],"adaptive":[72],"control":[74,82,165],"in":[75],"LLM":[77],"settings,":[78],"which":[79,122],"formulates":[80],"request-time":[81],"as":[83],"\\textit{partially":[85],"Markov":[87],"process}":[89],"capture":[91],"partial":[92],"observability":[93],"and":[94,158,160],"budget":[96],"coupling.":[97],"It":[98],"constructs":[99],"lightweight":[101],"verifiable":[102],"observation":[103],"channel":[104],"from":[105],"input-output":[107],"pair":[108],"by":[109,126],"aggregating":[110],"heterogeneous":[111],"quality":[112],"signals":[113],"into":[114],"belief":[116],"state":[117],"over":[118],"latent":[119],"reliability,":[121],"then":[124],"used":[125],"budget-aware":[128],"policy":[129],"return":[134],"output":[137],"trigger":[139],"higher-cost":[141],"pathway.":[143],"Experiments":[144],"on":[145],"diverse":[146],"tasks":[147],"show":[148],"that":[149],"\\textsc{Veroic}":[150],"achieves":[151],"improved":[152],"quality-cost":[153],"trade-offs,":[154],"risk":[156],"estimation":[157],"calibration,":[159],"more":[161],"robust":[162],"long-horizon":[163],"than":[166],"competitive":[167],"baselines.":[168]},"counts_by_year":[],"updated_date":"2026-05-02T06:10:54.344120","created_date":"2026-05-02T00:00:00"}