{"id":"https://openalex.org/W7165396967","doi":"https://doi.org/10.48550/arxiv.2606.19868","title":"A Systematic Evaluation of Black-Box Uncertainty Estimation Methods for Large Language Models","display_name":"A Systematic Evaluation of Black-Box Uncertainty Estimation Methods for Large Language Models","publication_year":2026,"publication_date":"2026-06-18","ids":{"openalex":"https://openalex.org/W7165396967","doi":"https://doi.org/10.48550/arxiv.2606.19868"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.19868","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.19868","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.19868","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139025716","display_name":"Jiayi Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Jiayi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138967612","display_name":"Xu-Yao Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Xu-Yao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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/T10028","display_name":"Topic Modeling","score":0.23680000007152557,"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/T10028","display_name":"Topic Modeling","score":0.23680000007152557,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.21310000121593475,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.09860000014305115,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.792900025844574},{"id":"https://openalex.org/keywords/trustworthiness","display_name":"Trustworthiness","score":0.611299991607666},{"id":"https://openalex.org/keywords/range","display_name":"Range (aeronautics)","score":0.4900999963283539},{"id":"https://openalex.org/keywords/estimation","display_name":"Estimation","score":0.4860000014305115},{"id":"https://openalex.org/keywords/empirical-research","display_name":"Empirical research","score":0.4056999981403351},{"id":"https://openalex.org/keywords/unified-model","display_name":"Unified Model","score":0.400299996137619}],"concepts":[{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.792900025844574},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7096999883651733},{"id":"https://openalex.org/C153701036","wikidata":"https://www.wikidata.org/wiki/Q659974","display_name":"Trustworthiness","level":2,"score":0.611299991607666},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.4900999963283539},{"id":"https://openalex.org/C96250715","wikidata":"https://www.wikidata.org/wiki/Q965330","display_name":"Estimation","level":2,"score":0.4860000014305115},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.47099998593330383},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.42980000376701355},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.40790000557899475},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.4056999981403351},{"id":"https://openalex.org/C45493050","wikidata":"https://www.wikidata.org/wiki/Q7884934","display_name":"Unified Model","level":2,"score":0.400299996137619},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.3456999957561493},{"id":"https://openalex.org/C2777617010","wikidata":"https://www.wikidata.org/wiki/Q18957","display_name":"Mainstream","level":2,"score":0.33480000495910645},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.329800009727478},{"id":"https://openalex.org/C37381756","wikidata":"https://www.wikidata.org/wiki/Q20203288","display_name":"Representativeness heuristic","level":2,"score":0.288100004196167},{"id":"https://openalex.org/C133199616","wikidata":"https://www.wikidata.org/wiki/Q25386885","display_name":"Empirical modelling","level":2,"score":0.25699999928474426}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.19868","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.19868","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.19868","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.19868","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":"Preprint"},"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":{"Although":[0],"large":[1],"language":[2],"models":[3,118],"(LLMs)":[4],"have":[5],"shown":[6],"strong":[7],"capabilities":[8],"across":[9,116,132],"a":[10,74,84,107,169],"wide":[11],"range":[12],"of":[13,87],"tasks,":[14],"their":[15],"outputs":[16],"often":[17],"remain":[18],"unreliable":[19],"and":[20,50,72,91,101,111,119,140,150,168,179],"may":[21],"contain":[22],"hallucinations,":[23],"making":[24,55],"uncertainty":[25,156],"estimation":[26],"(UE)":[27],"essential":[28],"for":[29,66,190,196],"building":[30],"trustworthy":[31],"LLMs.":[32,197],"In":[33],"practice,":[34],"many":[35],"mainstream":[36],"LLMs":[37,67],"are":[38,53,147],"only":[39],"accessible":[40],"through":[41],"restricted":[42],"APIs,":[43],"where":[44],"internal":[45],"signals":[46,157],"such":[47],"as":[48],"logits":[49],"hidden":[51],"states":[52],"unavailable,":[54],"black-box":[56,64,88,193],"UE":[57,65,89,194],"especially":[58],"important.":[59],"However,":[60],"existing":[61],"work":[62],"on":[63],"remains":[68],"fragmented":[69],"in":[70,143],"methodology":[71],"lacks":[73],"unified":[75,108,170],"empirical":[76,185],"comparison.":[77],"To":[78],"address":[79],"this":[80],"gap,":[81],"we":[82,173],"present":[83],"systematic":[85],"review":[86],"methods":[90,115,136,152,195],"organize":[92],"them":[93],"into":[94],"five":[95],"categories:":[96],"verbalization-based,":[97],"sampling-based,":[98],"explanation-based,":[99],"multi-agent,":[100],"hybrid":[102,151],"methods.":[103],"We":[104],"further":[105],"build":[106],"evaluation":[109,171],"framework":[110],"benchmark":[112,166],"24":[113],"representative":[114],"4":[117,120],"dataset":[121],"settings.":[122,134],"Our":[123],"results":[124],"show":[125],"that":[126,137,153],"no":[127],"single":[128],"method":[129],"consistently":[130],"dominates":[131],"all":[133],"Nevertheless,":[135],"reason":[138],"over":[139],"compare":[141],"candidates":[142],"the":[144,165],"answer":[145],"space":[146],"generally":[148],"effective,":[149],"combine":[154],"multiple":[155],"perform":[158],"well":[159],"under":[160],"most":[161],"conditions.":[162],"By":[163],"releasing":[164],"data":[167],"framework,":[172],"aim":[174],"to":[175],"facilitate":[176],"reproducible":[177],"comparisons":[178],"support":[180],"future":[181,192],"research,":[182],"while":[183],"our":[184],"findings":[186],"provide":[187],"practical":[188],"guidance":[189],"developing":[191]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-20T00:00:00"}